PRIMED Workflow for Group-Based Review

Introduction

This document contains the preliminary data analysis for the meta-analyses with dependent effects (PRIMED) in (Dalgaard et al. 2025). As we conduct separate analyses for reintegration (primary analysis) and mental health (secondary analysis) outcomes, we have divided the tabulation and visualization according to the two types of effect size estimates. In most cases, the main presentation of reintegration outcome data appears in the center column of the document, while the presentation of the mental health outcome data is shown in the right column. Where larger tables or visualizations are required, we have used tabsets to distinguish between reintegration and mental health analyses. To view the mental health presentation, select the ‘Mental health’ tab. To find the mental helath presentation, press on the ‘Mental health’ tab. In a few instances, reintegration and mental health outcomes are tabulated and visualized together to provide an overall view of the relationships between these two types of estimates. All packages that we have used to create this document, can be found in the next section.

Dalgaard, Nina T, Jakob K Jensen, Jasmin S Adada, Maya C Flensborg Jensen, Elizabeth Bengtsen, and Mikkel H Vembye. 2025. “Group-based community interventions to support the social reintegration of marginalized adults with mental illness: A systematic review and meta-analysis.” Campbell Systematic Reviews.

R packages

Below, we present the R package we use in this document. For exact R package versions, see the Session Information at the bottom of this document.

# Load packages -----------------------------------------------------------
library(knitr)
library(kableExtra)
library(skimr)
library(janitor)
library(tidyverse)
library(tidyr)
library(metafor)
library(clubSandwich)
library(fastDummies)  
library(ggrepel)
library(ggExtra)
library(ggridges)
library(MetBrewer)
library(GGally)
library(igraph)
library(fastDummies)
library(patchwork)
library(ggh4x)

Data manipulation - prepare data sets

In the following section, we create all the variables that are used in the main analyses of the review. Unfold the below code to see this exact manipulations.

Loading data

# Loading the needed data for analysis
group_based_dat <- readRDS("Group-based interventions data.RDS") |> 
  # The post-measurement of Empowerment Scale seems flawed for the study by 
  # Barbic et al. 2009 we therefore we exclude it from the analysis
  filter(!(authors == "Barbic et al." & test_name == "The Empowerment Scale")) |> 
  mutate(
    author_year = paste(authors, year),
    study = paste(authors, year),
    study = stringi::stri_trans_general(study, "Latin-ASCII"),
    study = fct_relevel(study, sort)
  ) |> 
 # Remove unused outcomes
 filter(!str_detect(analysis_plan, "Unused"))

Main variable manipulation

Unfold the below code, to find the primary data manipulation for the overall data, including both all reintegrational as well as mental health outcomes.

gb_dat <- 
  group_based_dat |> 
  # Exclude the two binary outcomes
  filter(variable_type != "Binary") |>
  mutate(
    es_id = 1:n(),
    esid = es_id,
    
    # Main covariate adjusted effect cluster adjusted 
    gt_pop = if_else(!is.na(gt_post), gt_post, NA_real_),
    gt_pop = if_else(!is.na(gt_DD), gt_DD, gt_pop),
    gt_pop = if_else(!is.na(gt_adj), gt_adj, gt_pop),
    gt_pop = if_else(!is.na(gt_reg), gt_reg, gt_pop),
    gt_pop = if_else(!is.na(gt_DD_pop), gt_DD_pop, gt_pop),
    gt_pop = if_else(!is.na(gt_adj_pop), gt_adj_pop, gt_pop),
    
    vgt_pop = if_else(!is.na(vgt_post), vgt_post, NA_real_),
    vgt_pop = if_else(!is.na(vgt_DD), vgt_DD, vgt_pop),
    vgt_pop = if_else(!is.na(vgt_adj), vgt_adj, vgt_pop),
    vgt_pop = if_else(!is.na(vgt_reg), vgt_reg, vgt_pop),
    vgt_pop = if_else(!is.na(vgt_DD_pop), vgt_DD_pop, vgt_pop),
    vgt_pop = if_else(!is.na(vgt_adj_pop), vgt_adj_pop, vgt_pop),
    
    Wgt_pop = if_else(!is.na(Wgt_post), Wgt_post, NA_real_),
    Wgt_pop = if_else(!is.na(Wgt_DD), Wgt_DD, Wgt_pop),
    Wgt_pop = if_else(!is.na(Wgt_adj), Wgt_adj, Wgt_pop),
    Wgt_pop = if_else(!is.na(Wgt_reg), Wgt_reg, Wgt_pop),
    Wgt_pop = if_else(!is.na(Wgt_DD_pop), Wgt_DD_pop, Wgt_pop),
    Wgt_pop = if_else(!is.na(Wgt_adj_pop), Wgt_adj_pop, Wgt_pop),
    
    # Main covariate adjusted effect cluster adjusted 
    gt = if_else(!is.na(gt_post), gt_post, NA_real_),
    gt = if_else(!is.na(gt_DD), gt_DD, gt),
    gt = if_else(!is.na(gt_adj), gt_adj, gt),
    gt = if_else(!is.na(gt_reg), gt_reg, gt),
    
    vgt = if_else(!is.na(vgt_post), vgt_post, NA_real_),
    vgt = if_else(!is.na(vgt_DD), vgt_DD, vgt),
    vgt = if_else(!is.na(vgt_adj), vgt_adj, vgt),
    vgt = if_else(!is.na(vgt_reg), vgt_reg, vgt),
    
    Wgt = if_else(!is.na(Wgt_post), Wgt_post, NA_real_),
    Wgt = if_else(!is.na(Wgt_DD), Wgt_DD, Wgt),
    Wgt = if_else(!is.na(Wgt_adj), Wgt_adj, Wgt),
    Wgt = if_else(!is.na(Wgt_reg), Wgt_reg, Wgt),
    
    # Hedges g posttest only, adjusted for clustering
    # Imputing covariate-adjusted estimate when posttest calculation was not possible
    gt_post = if_else(is.na(gt_post) & !is.na(gt_reg), gt_reg, gt_post),
    gt_post = if_else(is.na(gt_post) & !is.na(gt_adj), gt_adj, gt_post),
    gt_post = if_else(is.na(gt_post) & !is.na(gt_DD), gt_DD, gt_post),
    
    vgt_post = if_else(is.na(vgt_post) & !is.na(vgt_reg), vgt_reg, vgt_post),
    vgt_post = if_else(is.na(vgt_post) & !is.na(vgt_adj), vgt_adj, vgt_post),
    vgt_post = if_else(is.na(vgt_post) & !is.na(vgt_DD), vgt_DD, vgt_post),
    
    Wgt_post = if_else(is.na(Wgt_post) & !is.na(Wgt_reg), Wgt_reg, Wgt_post),
    Wgt_post = if_else(is.na(Wgt_post) & !is.na(Wgt_adj), Wgt_adj, Wgt_post),
    Wgt_post = if_else(is.na(Wgt_post) & !is.na(Wgt_DD), Wgt_DD, Wgt_post),
    
    # Hedges' g posttest only, not-adjusted for clustering
    # Imputing covariate-adjusted estimate when posttest calculation was not possible
    g_post = if_else(is.na(g_post) & !is.na(g_reg), g_reg, g_post),
    g_post = if_else(is.na(g_post) & !is.na(g_adj), g_adj, g_post),
    g_post = if_else(is.na(g_post) & !is.na(g_DD), g_DD, g_post),
    
    vg_post = if_else(is.na(vg_post) & !is.na(vg_reg), vg_reg, vg_post),
    vg_post = if_else(is.na(vg_post) & !is.na(vg_adj), vg_adj, vg_post),
    vg_post = if_else(is.na(vg_post) & !is.na(vg_DD), vg_DD, vg_post),
    
    Wg_post = if_else(is.na(Wg_post) & !is.na(Wg_reg), Wg_reg, Wg_post),
    Wg_post = if_else(is.na(Wg_post) & !is.na(Wg_adj), Wg_adj, Wg_post),
    Wg_post = if_else(is.na(Wg_post) & !is.na(Wg_DD), Wg_DD, Wg_post),
    
    # Cohen's d posttest only, not-adjusted for clustering
    # Imputing covariate-adjusted estimate when posttest calculation was not possible
    d_post = if_else(is.na(d_post) & !is.na(d_reg), d_reg, d_post),
    d_post = if_else(is.na(d_post) & !is.na(d_adj), d_adj, d_post),
    d_post = if_else(is.na(d_post) & !is.na(d_DD), d_DD, d_post),
    
    vd_post = if_else(is.na(vd_post) & !is.na(vd_reg), vd_reg, vd_post),
    vd_post = if_else(is.na(vd_post) & !is.na(vd_adj), vd_adj, vd_post),
    vd_post = if_else(is.na(vd_post) & !is.na(vd_DD), vd_DD, vd_post),
    
    Wd_post = if_else(is.na(Wd_post) & !is.na(Wd_reg), Wd_reg, Wd_post),
    Wd_post = if_else(is.na(Wd_post) & !is.na(Wd_adj), Wd_adj, Wd_post),
    Wd_post = if_else(is.na(Wd_post) & !is.na(Wd_DD), Wd_DD, Wd_post),
    
    
    # Covariate adjusted Hedges' g, not cluster adjusted 
    g = if_else(!is.na(g_post), g_post, NA_real_),
    g = if_else(!is.na(g_DD), g_DD, g),
    g = if_else(!is.na(g_adj), g_adj, g),
    g = if_else(!is.na(g_reg), g_reg, g),
    
    vg = if_else(!is.na(vg_post), vg_post, NA_real_),
    vg = if_else(!is.na(vg_DD), vg_DD, vg),
    vg = if_else(!is.na(vg_adj), vg_adj, vg),
    vg = if_else(!is.na(vg_reg), vg_reg, vg),
    
    Wg = if_else(!is.na(Wg_post), Wg_post, NA_real_),
    Wg = if_else(!is.na(Wg_DD), Wg_DD, Wg),
    Wg = if_else(!is.na(Wg_adj), Wg_adj, Wg),
    Wg = if_else(!is.na(Wg_reg), Wg_reg, Wg),
    
    # Covariate-adjusted version of Cohen's d, neither cluster nor small sample adjusted
    d = if_else(!is.na(d_post), d_post, NA_real_),
    d = if_else(!is.na(d_DD), d_DD, d),
    d = if_else(!is.na(d_adj), d_adj, d),
    d = if_else(!is.na(d_reg), d_reg, d),
    
    vd = if_else(!is.na(vd_post), vd_post, NA_real_),
    vd = if_else(!is.na(vd_DD), vd_DD, vd),
    vd = if_else(!is.na(vd_adj), vd_adj, vd),
    vd = if_else(!is.na(vd_reg), vd_reg, vd),
    
    Wd = if_else(!is.na(Wd_post), Wd_post, NA_real_),
    Wd = if_else(!is.na(Wd_DD), Wd_DD, Wd),
    Wd = if_else(!is.na(Wd_adj), Wd_adj, Wd),
    Wd = if_else(!is.na(Wd_reg), Wd_reg, Wd),
    
    inv_sample_size = (1/N_t + 1/N_c),
    
    posttest_only = if_else(ppcor_method == "Posttest only", "Posttest only ES", "Pre-posttest adj. ES", missing = "Pre-posttest adj. ES"),
    
    # ESS = round(4/vgt), # Using cluster bias corrected sampling variance
    
    studyid = if_else(authors == "Gonzalez & Prihoda", 500, studyid), 
    
    trt_id = if_else(str_detect(study, "Schaf") & trt_name == "Relapse Prevention Training", 2, trt_id),
    
    cnt = if_else(cnt == "USA", "US", cnt),
    
    design = if_else(design  == "QES-pretest", "QES", design),
    
    # MHV: Jeg synes ikke det er en god ide at ændre CRCT til RCT. Jeg vil gerne kunne se forskel.  
    # design = ifelse(design  == "CRCT", "RCT", design), 
    
    assessment = if_else(assessment == "Self  assesment", "Self assesment", assessment),
    
    randomization = if_else(randomization == "Simple Block Randomization", "Block randomized",
                            randomization),
    randomization = if_else(randomization == "Simple with permuted blocks, stratified by site",
                            "Stratified randomization", randomization),
    
    randomization = if_else(randomization == "Stratified?",
                            "Stratified randomization", randomization),
    
    randomization = if_else(is.na(randomization) & studyid == 102, 
                            "Stratified randomization", randomization),
    
    randomization = if_else(randomization == "Ratio and block randomized",
                            "Block randomized", randomization),
    
    randomization = if_else(randomization == "Block randomized stratified by site",
                            "Block randomized", randomization),
    
    randomization = if_else(randomization == "Resticted and adapted randomization (i.e. minimization)",
                            "Stratified randomization", randomization),
    
    randomization = if_else(randomization == "Unequal simple randomization",
                            "Block randomized", randomization),
    
    randomization = if_else(randomization == "Within-site basis and unequal allocation ratio",
                            "Stratified randomization", randomization),
    
    trt_type = if_else(trt_type == "Group-based  Cognitive Behavioral Therapy", 
                       "Group based Cognitive Behavioural Therapy", trt_type),
    trt_type = if_else(trt_type == "Group-based Cognitive Behavioral Therapy", 
                       "Group based Cognitive Behavioural Therapy", trt_type),
    trt_type = if_else(trt_type =="Group psychoeducation & Social skill training", 
                       "Group psychoeducation & Social Skill Training", trt_type),
    trt_type = if_else(trt_type =="Group psychoeducation", 
                       "Group Psychoeducation", trt_type),
    trt_type = if_else(trt_type =="Group psychoeducation & Social Skill Training", 
                       "Group Psychoeducation & Social Skill Training", trt_type),
    trt_type = if_else(trt_type =="Education and Illness Management", 
                       "Group Psychoeducation & Social Skill Training", trt_type),
    trt_type = if_else(trt_type =="Illness management", 
                       "Illness Management", trt_type),
    
    sample_factors = if_else(sample_factors =="Older with depression and anxiety", 
                             "Mixed", sample_factors), 
    sample_factors = if_else(sample_factors =="All suffered from severe mental illness", 
                             "Shared Social problem(s)/challenge(s)", sample_factors),
    sample_factors = if_else(sample_factors =="Shared origin and psychological distress", 
                             "Mixed", sample_factors),
    sample_factors = if_else(sample_factors =="persons with major psychiatric problems", 
                             "Shared Social problem(s)/challenge(s)", sample_factors),
    
    analysis_plan = if_else(analysis_plan == "Unused", "Unused outcomes", analysis_plan), 
    analysis_plan = case_match(
      analysis_plan,
      "Hope, Empowerment & Self-efficacy" ~ "Hope, empowerment & self-efficacy",
      "Wellbeing and Quality of Life" ~ "Wellbeing and quality of life",
      "All mental health outcomes" ~ "General mental health",
      "All mental health outcomes/Anxiety" ~ "Anxiety",
      "All mental health outcomes/Depression" ~ "Depression",
      "All mental health outcomes/Symptoms of psychosis" ~ "Symptoms of psychosis",
      "All mental health outcomes/Negative symptoms" ~ "Symptoms of psychosis",
      .default = analysis_plan
    ),
    
    test_type = if_else(test_type == "Clinical administered", "Clinician-rated measure", test_type),
    test_type = if_else(test_type == "Clinical interviews", "Clinician-rated measure", test_type),
    test_type = if_else(test_type == "Self report", "Self-reported", test_type),
    test_type = if_else(test_type == "Self report through clinical interview", "Self-reported", test_type),
    test_type = if_else(test_type == "Self-reported via diagnostic interview", "Self-reported", test_type), 
    
    measure_type = if_else(measure_type == "Pre-post with controls", "Post-intervention", measure_type),
    
    cluster_treatment = if_else(cluster_treatment == "Hierarchical mixed models", "Mixed-model", cluster_treatment),
    
    cluster_treatment = if_else(cluster_treatment == "Multilevel analysis and clustered standard errors", 
                                "Multilevel analysis", cluster_treatment),
    
    rob_tool = if_else(rob_tool == "Rob2", "RoB2", rob_tool),
    rob_tool = if_else(rob_tool == "Rob2 CRCT", "RoB2", rob_tool),
    
    analysis_strategy = if_else(str_detect(study, "Michalak"), "ITT", analysis_strategy),
    
    conventional = if_else(protocol != "Yes", 1, 0),
    # Smith et al. 2021 drew on a retropective protocol
    conventional = if_else(study == "Smith et al. 2021", 1, conventional),
    prereg_chr = if_else(conventional == 0, "Preregistered", "Not preregistered"),
    
    # For publication/selection/small study bias testing
    Wse = sqrt(Wgt),
    t_i = gt/sqrt(Wgt),
    
    ESS_t = round(N_t/(1+(avg_cl_size - 1)*0.1)),
    ESS_c = round(N_c/(1+(avg_cl_size - 1)*0.1)),
    ESS_total = round(N_total/(1+(avg_cl_size - 1)*0.1)),
    
    outcome_construct = case_match(
      analysis_plan,
      # Mental health outcomes
      c("General mental health", "Anxiety",
        "Depression", "Symptoms of psychosis") ~ "Mental health outcome",
      .default = "Reintegrational outcome"
    ),
    
    # Changing to numeric vectors
    across(c(age_mean_sample:male_pct_t, sessions_per_week), ~as.numeric(.x)),
    
    # Make weighted mean weighted by the group sample size 
    age_mean = if_else(
      is.na(age_mean_sample), 
      (age_mean_t*N_t + age_mean_c*N_c)/(N_t + N_c), 
      age_mean_sample
    ),
    
    male_pct = if_else(
      is.na(male_pct_sample),
      (male_pct_t*N_t + male_pct_c*N_c)/(N_t + N_c),
      male_pct_sample
    ),
    
    # duration_weeks has extract errors
    duration_in_weeks = time_from_baseline_weeks - time_after_end_intervention_weeks,
    total_number_of_sessions = round(sessions_per_week * duration_in_weeks),
    
    CBT_int = if_else(trt_group == "group-based CBT", "CBT", "Other"), 
    
    QES_design = if_else(design == "QES", "QES", "RCT"),
    
    overall_rob = case_match(
      Overall, 
      c("Serious", "High") ~ "Serious/High",
      c("Some concerns", "Moderate") ~ "Some concerns/Moderate",
      .default = "Low"
    ),
    
    overall_rob = factor(overall_rob, levels = c("Low", "Some concerns/Moderate", "Serious/High")),
    
    across(schizophrenia_or_primary_psychotic_disorder:dissociative_identity_disorder, ~ replace_na(.x, 0)),
    handle_multilevel = as.numeric(if_any(adj_fct_DD:adj_fct_reg, ~ grepl("gamma", .))),
    handle_multilevel = if_else(handle_multilevel == 1, "Yes", "No")
    
 ) |> 
  rowwise() |> 
  mutate(
    diagnosis = {
      col_ones <- names(
        across(.cols = schizophrenia_or_primary_psychotic_disorder:dissociative_identity_disorder)
        )[unlist(
          c_across(schizophrenia_or_primary_psychotic_disorder:dissociative_identity_disorder)
          ) == 1]
      n_ones <- length(col_ones)
      if (n_ones == 1) col_ones
      else if (n_ones > 1) "mixed"
    }
  ) |> 
  ungroup() |> 
  mutate(
    schizophrenia = if_else(str_detect(diagnosis, "schizophrenia"), "Schizophrenia", "Other")
  ) |> 
  mutate(
    # Used to remove ITT outcomes from Cano-Vindel et al. 2021 and Craigie & Nathan 2009 
    n_analysis_strategies = n_distinct(analysis_strategy),
    .by = study
  ) |> 
  # Removing ITT analyses from Cano-Vindel et al. 2021 and Craigie & Nathan 2009 
  filter(!c(str_detect(study, "Cano|Craigie|Woj") & analysis_strategy == "ITT")) 

time_vcalc <- c(
  # Acarturk
  1,2,1,2,1,2,
  # Barbic
  1,1,1,
  # Bond
  1,1, 
  # Baekkelund, 
  1,1,2,2,
  # Cano-Vindel
  rep(c(1:4), each = 10),
  # Craigie & Nathan
  1,1,1,
  # Crawford (multi-treatment)
  rep(c(1,2), each = 2, 4),
  # Druss 2010
  1,1,
  #Druss 2018
  1,1,1,2,2,2,
  # Dyck
  1, 
  # Gestel-Timmermans
  rep(c(1,2), 5),
  # Gatz
  rep(1,5),
  # Gonzalaez
  rep(1,2),
  # Gordon 
  rep(1,4),
  # Gutman
  rep(1,2),
  # Hagen
  rep(1,4),
  # Haslem
  rep(1,3),
  # Hilden
  rep(1,7),
  # Himle
  rep(c(1,2), each = 6),
  #Jacob
  rep(c(1,2), each = 5),
  #James
  rep(1,5),
  # Kanie
  rep(1,4),
  # Lim
  rep(1,6),
  # Llyod-Evans
  rep(1,4),
  # Madigan
  rep(c(1,2), each = 6),
  # McCay 2006
  rep(1,4),
  # McCay 2007
  rep(1,3),
  # Michalak (multi-treatment)
  rep(1,14),
  # Morley 
  rep(1,4),
  # Morton
  rep(1,5),
  # Patterson
  rep(c(1,2), each = 5),
  # Popolo
  rep(1,2),
  # Rabenstein
  rep(1,3),
  # Rosenblum 
  rep(1,4),
  # Russinova 
  rep(c(1,2), each = 5),
  # Rusch
  rep(c(1,2), each = 6),
  # Sacks
  rep(1,5),
  # Sajatovic
  rep(c(1:3), each = 3),
  # Saloheimo
  1, 
  # Schrank
  rep(1,6),
  # Schafer (multi-treatment)
  rep(c(2,3,1), each = 2, 6),
  # Somers
  rep(1,5),
  # Tjaden
  rep(c(1,2), each = 4),
  # Valiente 
  rep(1,16),
  # Volpe
  rep(1,4),
  # Wojtalik 
  rep(1,2),
  #Wuthrich
  rep(1,4),
  # Smith 
  rep(1,3)
) 

gb_dat$time <- time_vcalc

Creating primary and secondary data

Below, we separate the data by reintegrational (primary analysis) and mental health outcomes (secondary analyses)

Reintegration data

A general overview of the main data, we use for analyses of reintegrational outcomes can be found in the scroll box below.

reintegration_dat <- 
  gb_dat |> 
  filter(outcome_construct == "Reintegrational outcome") 

saveRDS(reintegration_dat, file = "reintegration_dat.rds")

reint_overview <- 
  reintegration_dat |> 
  select(
    study, eppi_id, esid, N_t, N_c, N_total, inv_sample_size, gt_pop, vgt_pop, Wgt_pop, gt, vgt, Wgt, Wse, 
    prereg_chr, conventional, analysis_plan, Overall, D5, D7, timing
  )

reint_overview |> 
  mutate(
    p_val = 2 * ( 1 - pnorm( abs(gt_pop) / sqrt(vgt_pop) ) )
  ) |> 
  select(
    `Authors (year)` = study, N_t, N_c,
    `Outcome construct` = analysis_plan, gt_pop, vgt_pop, Wgt, Wse, p_val, 
    `No protocol` = conventional, `Overall RoB` = Overall
  ) |> 
  kable(digits=3)  |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "600px", fixed_thead = TRUE)

Table 1: Data with reintegration outcomes.

Authors (year)	N_t	N_c	Outcome construct	gt_pop	vgt_pop	Wgt	Wse	p_val	No protocol	Overall RoB
Acarturk et al. 2022	24	22	Wellbeing and quality of life	0.314	0.196	0.195	0.442	0.478	0	Some concerns
Acarturk et al. 2022	24	22	Wellbeing and quality of life	-0.103	0.091	0.091	0.301	0.733	0	High
Barbic et al. 2009	16	17	Hope, empowerment & self-efficacy	0.443	0.162	0.159	0.399	0.272	1	Some concerns
Barbic et al. 2009	16	17	Wellbeing and quality of life	-0.069	0.159	0.159	0.399	0.863	1	Some concerns
Barbic et al. 2009	16	17	Hope, empowerment & self-efficacy	0.548	0.164	0.159	0.399	0.176	1	Some concerns
Bond et al. 2015	43	42	Hope, empowerment & self-efficacy	0.000	0.094	0.094	0.306	1.000	1	Some concerns
Bond et al. 2015	43	41	Psychiatric hospitalization	0.309	0.095	0.094	0.307	0.316	1	Some concerns
Baekkelund et al. 2022	29	30	Social functioning (degree of impairment)	-0.082	0.078	0.082	0.286	0.769	0	Low
Baekkelund et al. 2022	29	30	Social functioning (degree of impairment)	0.142	0.084	0.088	0.297	0.624	0	Low
Cano-Vindel et al. 2021	315	316	Social functioning (degree of impairment)	0.259	0.006	0.006	0.080	0.001	0	Low
Cano-Vindel et al. 2021	315	316	Social functioning (degree of impairment)	0.428	0.005	0.005	0.072	0.000	0	Low
Cano-Vindel et al. 2021	315	316	Social functioning (degree of impairment)	0.439	0.007	0.007	0.082	0.000	0	Low
Cano-Vindel et al. 2021	315	316	Wellbeing and quality of life	0.565	0.006	0.006	0.076	0.000	0	Low
Cano-Vindel et al. 2021	315	316	Wellbeing and quality of life	0.593	0.005	0.005	0.069	0.000	0	Low
Cano-Vindel et al. 2021	315	316	Wellbeing and quality of life	0.305	0.005	0.005	0.069	0.000	0	Low
Cano-Vindel et al. 2021	315	316	Wellbeing and quality of life	0.389	0.007	0.007	0.083	0.000	0	Low
Cano-Vindel et al. 2021	273	238	Social functioning (degree of impairment)	0.066	0.008	0.008	0.088	0.450	0	Low
Cano-Vindel et al. 2021	273	238	Social functioning (degree of impairment)	0.128	0.006	0.006	0.079	0.105	0	Low
Cano-Vindel et al. 2021	273	238	Social functioning (degree of impairment)	0.224	0.008	0.008	0.090	0.013	0	Low
Cano-Vindel et al. 2021	273	238	Wellbeing and quality of life	0.301	0.007	0.007	0.083	0.000	0	Low
Cano-Vindel et al. 2021	273	238	Wellbeing and quality of life	0.308	0.006	0.006	0.076	0.000	0	Low
Cano-Vindel et al. 2021	273	238	Wellbeing and quality of life	0.211	0.006	0.006	0.076	0.005	0	Low
Cano-Vindel et al. 2021	273	238	Wellbeing and quality of life	0.194	0.008	0.008	0.091	0.033	0	Low
Cano-Vindel et al. 2021	229	204	Social functioning (degree of impairment)	0.342	0.009	0.009	0.095	0.000	0	Low
Cano-Vindel et al. 2021	229	204	Social functioning (degree of impairment)	0.357	0.008	0.007	0.086	0.000	0	Low
Cano-Vindel et al. 2021	229	204	Social functioning (degree of impairment)	0.403	0.010	0.010	0.098	0.000	0	Low
Cano-Vindel et al. 2021	229	204	Wellbeing and quality of life	0.381	0.008	0.008	0.091	0.000	0	Low
Cano-Vindel et al. 2021	229	204	Wellbeing and quality of life	0.232	0.007	0.007	0.082	0.005	0	Low
Cano-Vindel et al. 2021	229	204	Wellbeing and quality of life	0.083	0.007	0.007	0.082	0.313	0	Low
Cano-Vindel et al. 2021	229	204	Wellbeing and quality of life	0.179	0.010	0.010	0.099	0.072	0	Low
Cano-Vindel et al. 2021	208	180	Social functioning (degree of impairment)	0.455	0.010	0.010	0.101	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Social functioning (degree of impairment)	0.457	0.008	0.008	0.091	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Social functioning (degree of impairment)	0.512	0.011	0.011	0.103	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Wellbeing and quality of life	0.598	0.010	0.009	0.096	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Wellbeing and quality of life	0.410	0.008	0.008	0.087	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Wellbeing and quality of life	0.281	0.008	0.008	0.087	0.001	0	Low
Cano-Vindel et al. 2021	208	180	Wellbeing and quality of life	0.245	0.011	0.011	0.105	0.020	0	Low
Craigie & Nathan 2009	157	77	Wellbeing and quality of life	-0.247	0.026	0.026	0.162	0.129	1	Serious
Crawford et al. 2012	121	121	Social functioning (degree of impairment)	-0.022	0.024	0.025	0.159	0.885	0	Low
Crawford et al. 2012	119	121	Social functioning (degree of impairment)	-0.052	0.024	0.026	0.160	0.739	0	Low
Crawford et al. 2012	121	117	Social functioning (degree of impairment)	-0.039	0.019	0.020	0.141	0.779	0	Low
Crawford et al. 2012	117	117	Social functioning (degree of impairment)	-0.085	0.019	0.020	0.142	0.542	0	Low
Crawford et al. 2012	121	121	Social functioning (degree of impairment)	-0.265	0.022	0.021	0.147	0.071	0	Low
Crawford et al. 2012	119	121	Social functioning (degree of impairment)	-0.137	0.022	0.022	0.147	0.352	0	Low
Crawford et al. 2012	121	117	Social functioning (degree of impairment)	-0.209	0.022	0.022	0.147	0.158	0	Low
Crawford et al. 2012	117	117	Social functioning (degree of impairment)	-0.081	0.022	0.022	0.149	0.588	0	Low
Crawford et al. 2012	121	121	Wellbeing and quality of life	0.203	0.022	0.021	0.147	0.168	0	Low
Crawford et al. 2012	119	121	Wellbeing and quality of life	0.074	0.022	0.022	0.147	0.616	0	Low
Crawford et al. 2012	121	117	Wellbeing and quality of life	0.177	0.022	0.022	0.147	0.229	0	Low
Crawford et al. 2012	117	117	Wellbeing and quality of life	0.166	0.022	0.022	0.149	0.266	0	Low
Druss et al. 2010	41	39	Wellbeing and quality of life	0.207	0.065	0.065	0.254	0.417	0	High
Druss et al. 2010	41	39	Wellbeing and quality of life	0.035	0.065	0.065	0.254	0.890	0	High
Druss et al. 2018	198	202	Hope, empowerment & self-efficacy	0.204	0.013	0.013	0.114	0.074	0	Some concerns
Druss et al. 2018	198	202	Wellbeing and quality of life	0.055	0.013	0.013	0.114	0.630	0	Some concerns
Druss et al. 2018	198	202	Wellbeing and quality of life	0.019	0.013	0.013	0.114	0.866	0	Some concerns
Druss et al. 2018	198	202	Hope, empowerment & self-efficacy	0.113	0.013	0.013	0.114	0.321	0	Some concerns
Druss et al. 2018	198	202	Wellbeing and quality of life	0.102	0.013	0.013	0.114	0.373	0	Some concerns
Druss et al. 2018	198	202	Wellbeing and quality of life	0.169	0.013	0.013	0.114	0.140	0	Some concerns
Gestel-Timmermans et al. 2012	136	117	Hope, empowerment & self-efficacy	0.193	0.014	0.014	0.118	0.103	0	High
Gestel-Timmermans et al. 2012	121	99	Hope, empowerment & self-efficacy	0.156	0.014	0.014	0.120	0.192	0	High
Gestel-Timmermans et al. 2012	132	118	Hope, empowerment & self-efficacy	0.122	0.014	0.014	0.117	0.298	0	High
Gestel-Timmermans et al. 2012	120	97	Hope, empowerment & self-efficacy	0.348	0.017	0.017	0.129	0.008	0	High
Gestel-Timmermans et al. 2012	138	122	Loneliness	-0.196	0.014	0.014	0.116	0.093	0	High
Gestel-Timmermans et al. 2012	125	102	Loneliness	-0.039	0.014	0.014	0.117	0.740	0	High
Gestel-Timmermans et al. 2012	124	114	Wellbeing and quality of life	-0.045	0.011	0.011	0.106	0.675	0	High
Gestel-Timmermans et al. 2012	111	97	Wellbeing and quality of life	-0.039	0.012	0.012	0.111	0.727	0	High
Gatz et al. 2007	136	177	Alcohol and drug abuse/misuse	0.094	0.023	0.022	0.150	0.531	1	Serious
Gatz et al. 2007	135	176	Alcohol and drug abuse/misuse	0.204	0.046	0.046	0.214	0.341	1	Serious
Gatz et al. 2007	134	173	Hope, empowerment & self-efficacy	0.241	0.019	0.019	0.138	0.082	1	Serious
Gonzalez & Prihoda 2007	8	9	Social functioning (degree of impairment)	0.206	0.287	0.189	0.434	0.701	1	Serious
Gordon et al. 2018	21	15	Wellbeing and quality of life	0.258	0.142	0.141	0.375	0.494	1	Some concerns
Gordon et al. 2018	21	15	Social functioning (degree of impairment)	-0.388	0.143	0.141	0.375	0.305	1	Some concerns
Gordon et al. 2018	21	15	Social functioning (degree of impairment)	0.183	0.141	0.141	0.375	0.627	1	Some concerns
Gutman et al. 2019	10	10	Wellbeing and quality of life	0.501	0.033	0.026	0.162	0.006	1	Serious
Hagen et al. 2005	14	17	Social functioning (degree of impairment)	-0.039	0.174	0.174	0.418	0.925	1	High
Haslam et al. 2019	66	54	Loneliness	-0.246	0.006	0.005	0.073	0.001	0	High
Hilden et al. 2021	23	12	Physical health	0.189	0.138	0.140	0.375	0.612	0	High
Hilden et al. 2021	23	12	Alcohol and drug abuse/misuse	0.070	0.140	0.140	0.375	0.852	0	High
Hilden et al. 2021	23	12	Social functioning (degree of impairment)	-0.283	0.142	0.140	0.375	0.452	0	High
Hilden et al. 2021	23	12	Social functioning (degree of impairment)	0.434	0.143	0.140	0.375	0.252	0	High
Hilden et al. 2021	23	12	Social functioning (degree of impairment)	0.036	0.140	0.140	0.375	0.924	0	High
Himle et al. 2014	29	29	Physical health	0.507	0.083	0.083	0.288	0.079	0	Some concerns
Himle et al. 2014	29	29	Social functioning (degree of impairment)	0.464	0.085	0.083	0.288	0.111	0	Some concerns
Himle et al. 2014	29	29	Physical health	0.523	0.083	0.083	0.288	0.070	0	Some concerns
Himle et al. 2014	29	29	Social functioning (degree of impairment)	0.819	0.089	0.083	0.288	0.006	0	Some concerns
Jacob et al. 2010	19	24	Self-esteem	0.495	0.120	0.116	0.341	0.152	1	Serious
Jacob et al. 2010	19	24	Self-esteem	0.442	0.119	0.116	0.341	0.200	1	Serious
Jacob et al. 2010	19	24	Self-esteem	0.133	0.117	0.116	0.341	0.697	1	Serious
Jacob et al. 2010	19	24	Self-esteem	0.053	0.116	0.116	0.341	0.876	1	Serious
Jacob et al. 2010	19	24	Self-esteem	0.392	0.118	0.116	0.341	0.255	1	Serious
Jacob et al. 2010	19	24	Self-esteem	0.433	0.119	0.116	0.341	0.209	1	Serious
Jacob et al. 2010	19	24	Self-esteem	0.222	0.117	0.116	0.341	0.516	1	Serious
Jacob et al. 2010	19	24	Self-esteem	-0.121	0.117	0.116	0.341	0.724	1	Serious
James et al. 2004	29	29	Alcohol and drug abuse/misuse	1.120	0.094	0.083	0.288	0.000	1	Some concerns
James et al. 2004	29	29	Alcohol and drug abuse/misuse	0.281	0.083	0.083	0.288	0.331	1	Some concerns
James et al. 2004	28	29	Alcohol and drug abuse/misuse	0.538	0.087	0.085	0.291	0.069	1	Some concerns
Kanie et al. 2019	32	29	Social functioning (degree of impairment)	-0.177	0.075	0.078	0.279	0.517	0	Low
Kanie et al. 2019	32	29	Social functioning (degree of impairment)	0.037	0.078	0.078	0.279	0.896	0	Low
Lim et al. 2020	18	21	Wellbeing and quality of life	0.862	0.148	0.137	0.371	0.025	1	Serious
Lloyd-Evans et al. 2020	25	10	Loneliness	0.103	0.158	0.158	0.398	0.796	0	Low
Lloyd-Evans et al. 2020	25	10	Wellbeing and quality of life	0.214	0.159	0.158	0.398	0.591	0	Low
Madigan et al. 2013	36	18	Alcohol and drug abuse/misuse	0.024	0.097	0.097	0.312	0.940	1	Some concerns
Madigan et al. 2013	39	19	Social functioning (degree of impairment)	0.038	0.087	0.091	0.302	0.898	1	Some concerns
Madigan et al. 2013	34	15	Wellbeing and quality of life	0.490	0.113	0.110	0.332	0.144	1	Some concerns
Madigan et al. 2013	28	14	Alcohol and drug abuse/misuse	0.049	0.125	0.125	0.354	0.891	1	Some concerns
Madigan et al. 2013	31	16	Social functioning (degree of impairment)	0.008	0.106	0.111	0.333	0.982	1	Some concerns
Madigan et al. 2013	34	14	Wellbeing and quality of life	0.677	0.119	0.114	0.338	0.050	1	Some concerns
McCay et al. 2006	26	14	Self-esteem	0.521	0.052	0.048	0.219	0.022	1	Serious
McCay et al. 2006	26	14	Wellbeing and quality of life	0.129	0.109	0.108	0.329	0.694	1	Serious
McCay et al. 2007	29	18	Self-esteem	0.386	0.018	0.016	0.128	0.004	1	High
McCay et al. 2007	29	18	Wellbeing and quality of life	0.509	0.019	0.016	0.126	0.000	1	High
McCay et al. 2007	29	18	Hope, empowerment & self-efficacy	0.400	0.021	0.019	0.137	0.005	1	High
Michalak et al. 2015	36	35	Wellbeing and quality of life	0.299	0.055	0.054	0.233	0.203	0	Some concerns
Michalak et al. 2015	35	35	Wellbeing and quality of life	0.349	0.033	0.032	0.180	0.056	0	Some concerns
Michalak et al. 2015	36	35	Wellbeing and quality of life	0.115	0.032	0.031	0.177	0.519	0	Some concerns
Michalak et al. 2015	35	35	Wellbeing and quality of life	0.329	0.037	0.036	0.190	0.087	0	Some concerns
Michalak et al. 2015	36	35	Wellbeing and quality of life	0.472	0.054	0.052	0.229	0.043	0	Some concerns
Michalak et al. 2015	35	35	Wellbeing and quality of life	0.410	0.044	0.043	0.207	0.052	0	Some concerns
Michalak et al. 2015	36	35	Wellbeing and quality of life	0.134	0.039	0.038	0.196	0.495	0	Some concerns
Michalak et al. 2015	35	35	Wellbeing and quality of life	0.270	0.035	0.035	0.186	0.151	0	Some concerns
Michalak et al. 2015	36	35	Social functioning (degree of impairment)	0.567	0.055	0.052	0.229	0.015	0	Some concerns
Michalak et al. 2015	35	35	Social functioning (degree of impairment)	0.389	0.045	0.044	0.209	0.066	0	Some concerns
Morley et al. 2014	122	63	Hope, empowerment & self-efficacy	0.296	0.028	0.027	0.166	0.075	1	Some concerns
Morton et al. 2012	21	20	Hope, empowerment & self-efficacy	0.985	0.091	0.078	0.279	0.001	1	Some concerns
Patterson et al. 2003	16	16	Wellbeing and quality of life	0.109	0.163	0.162	0.403	0.788	1	Some concerns
Patterson et al. 2003	16	16	Wellbeing and quality of life	-0.268	0.164	0.162	0.403	0.508	1	Some concerns
Rabenstein et al. 2016	153	148	Wellbeing and quality of life	0.415	0.018	0.018	0.133	0.002	1	Moderate
Rosenblum et al. 2014	91	70	Alcohol and drug abuse/misuse	0.129	0.032	0.032	0.178	0.468	0	Some concerns
Rosenblum et al. 2014	91	70	Alcohol and drug abuse/misuse	0.325	0.032	0.032	0.178	0.069	0	Some concerns
Rosenblum et al. 2014	91	70	Alcohol and drug abuse/misuse	0.168	0.032	0.032	0.178	0.345	0	Some concerns
Rosenblum et al. 2014	91	70	Alcohol and drug abuse/misuse	0.178	0.032	0.032	0.178	0.318	0	Some concerns
Russinova et al. 2018	22	26	Hope, empowerment & self-efficacy	0.131	0.112	0.112	0.334	0.696	1	Some concerns
Russinova et al. 2018	22	26	Self-esteem	0.716	0.118	0.112	0.334	0.037	1	Some concerns
Russinova et al. 2018	22	26	Hope, empowerment & self-efficacy	0.101	0.112	0.112	0.334	0.762	1	Some concerns
Russinova et al. 2018	22	26	Employment	0.296	0.113	0.112	0.334	0.378	1	Some concerns
Russinova et al. 2018	21	24	Hope, empowerment & self-efficacy	-0.117	0.119	0.118	0.344	0.733	1	Some concerns
Russinova et al. 2018	22	25	Hope, empowerment & self-efficacy	0.181	0.114	0.113	0.337	0.591	1	Some concerns
Russinova et al. 2018	22	25	Self-esteem	0.225	0.114	0.113	0.337	0.505	1	Some concerns
Russinova et al. 2018	22	25	Hope, empowerment & self-efficacy	-0.113	0.113	0.113	0.337	0.737	1	Some concerns
Russinova et al. 2018	22	25	Employment	0.030	0.113	0.113	0.337	0.930	1	Some concerns
Russinova et al. 2018	21	24	Hope, empowerment & self-efficacy	0.000	0.118	0.118	0.344	1.000	1	Some concerns
Rusch et al. 2019	18	17	Hope, empowerment & self-efficacy	0.333	0.104	0.102	0.319	0.301	0	Some concerns
Rusch et al. 2019	18	17	Hope, empowerment & self-efficacy	-0.094	0.057	0.057	0.239	0.694	0	Some concerns
Rusch et al. 2019	18	17	Hope, empowerment & self-efficacy	0.650	0.125	0.118	0.344	0.066	0	Some concerns
Rusch et al. 2019	18	17	Self-esteem	0.511	0.035	0.031	0.177	0.006	0	Some concerns
Rusch et al. 2019	18	17	Hope, empowerment & self-efficacy	0.522	0.107	0.103	0.321	0.111	0	Some concerns
Rusch et al. 2019	20	13	Hope, empowerment & self-efficacy	0.196	0.155	0.154	0.393	0.618	0	Some concerns
Rusch et al. 2019	20	13	Hope, empowerment & self-efficacy	-0.258	0.155	0.154	0.393	0.512	0	Some concerns
Rusch et al. 2019	20	13	Hope, empowerment & self-efficacy	0.772	0.391	0.381	0.617	0.217	0	Some concerns
Rusch et al. 2019	20	13	Self-esteem	0.338	0.094	0.092	0.304	0.271	0	Some concerns
Rusch et al. 2019	20	13	Hope, empowerment & self-efficacy	0.724	0.274	0.266	0.515	0.167	0	Some concerns
Sacks et al. 2011	38	38	Wellbeing and quality of life	0.074	0.068	0.068	0.262	0.779	1	Some concerns
Sacks et al. 2011	38	38	Wellbeing and quality of life	-0.269	0.069	0.068	0.262	0.305	1	Some concerns
Sacks et al. 2011	38	38	Wellbeing and quality of life	0.953	0.075	0.068	0.262	0.000	1	Some concerns
Sajatovic et al. 2009	61	61	Social functioning (degree of impairment)	0.220	0.041	0.041	0.202	0.279	1	Some concerns
Sajatovic et al. 2009	46	53	Social functioning (degree of impairment)	0.086	0.052	0.052	0.228	0.706	1	Some concerns
Sajatovic et al. 2009	40	39	Social functioning (degree of impairment)	0.059	0.063	0.063	0.251	0.815	1	Some concerns
Schrank et al. 2016	43	41	Wellbeing and quality of life	-0.104	0.062	0.061	0.248	0.675	0	Low
Schrank et al. 2016	43	41	Wellbeing and quality of life	0.445	0.063	0.061	0.248	0.076	0	Low
Schrank et al. 2016	43	41	Hope, empowerment & self-efficacy	-0.341	0.062	0.061	0.248	0.171	0	Low
Schrank et al. 2016	43	41	Hope, empowerment & self-efficacy	0.021	0.061	0.061	0.248	0.932	0	Low
Schrank et al. 2016	43	41	Hope, empowerment & self-efficacy	0.148	0.062	0.061	0.248	0.550	0	Low
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	0.144	0.023	0.022	0.150	0.339	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	0.151	0.023	0.023	0.152	0.321	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	0.346	0.023	0.022	0.150	0.022	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	-0.036	0.023	0.023	0.152	0.813	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	0.223	0.023	0.022	0.150	0.138	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	0.148	0.023	0.023	0.152	0.329	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	0.088	0.022	0.022	0.150	0.558	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	0.000	0.023	0.023	0.152	1.000	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	0.369	0.023	0.022	0.150	0.015	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	0.184	0.023	0.023	0.152	0.226	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	0.264	0.023	0.022	0.150	0.079	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	0.084	0.023	0.023	0.152	0.579	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	-0.306	0.023	0.022	0.150	0.042	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	-0.074	0.023	0.023	0.152	0.625	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	-0.495	0.023	0.022	0.150	0.001	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	-0.249	0.023	0.023	0.152	0.101	0	Some concerns
Schafer et al. 2019	115	117	Alcohol and drug abuse/misuse	-0.249	0.023	0.022	0.150	0.097	0	Some concerns
Schafer et al. 2019	111	117	Alcohol and drug abuse/misuse	-0.080	0.023	0.023	0.152	0.600	0	Some concerns
Somers et al. 2017	90	100	Social functioning (degree of impairment)	0.252	0.034	0.034	0.184	0.172	0	Low
Somers et al. 2017	90	100	Social functioning (degree of impairment)	0.661	0.044	0.043	0.208	0.002	0	Low
Somers et al. 2017	90	100	Alcohol and drug abuse/misuse	-0.143	0.041	0.041	0.203	0.480	0	Low
Somers et al. 2017	90	100	Wellbeing and quality of life	0.250	0.034	0.033	0.183	0.172	0	Low
Tjaden et al. 2021	70	67	Hope, empowerment & self-efficacy	0.353	0.038	0.038	0.194	0.071	0	Low
Tjaden et al. 2021	70	67	Wellbeing and quality of life	0.082	0.038	0.038	0.194	0.672	0	Low
Tjaden et al. 2021	70	67	Social functioning (degree of impairment)	-0.213	0.036	0.038	0.194	0.262	0	Low
Tjaden et al. 2021	63	58	Hope, empowerment & self-efficacy	0.532	0.044	0.042	0.206	0.011	0	Low
Tjaden et al. 2021	63	58	Wellbeing and quality of life	-0.032	0.042	0.042	0.206	0.875	0	Low
Tjaden et al. 2021	63	58	Social functioning (degree of impairment)	0.008	0.041	0.042	0.206	0.970	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	0.150	0.048	0.047	0.218	0.492	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	0.104	0.048	0.047	0.218	0.632	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	0.138	0.048	0.047	0.218	0.526	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	0.325	0.048	0.047	0.218	0.137	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	0.136	0.048	0.047	0.218	0.534	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	0.028	0.047	0.047	0.218	0.897	0	Low
Valiente et al. 2022	52	61	Wellbeing and quality of life	-0.172	0.048	0.047	0.218	0.430	0	Low
Volpe et al. 2015	21	20	Social functioning (degree of impairment)	0.092	0.126	0.126	0.355	0.795	1	Some concerns
Volpe et al. 2015	21	20	Social functioning (degree of impairment)	1.355	0.150	0.126	0.355	0.000	1	Some concerns
Wojtalik et al. 2022	26	23	Social functioning (degree of impairment)	0.105	0.089	0.089	0.299	0.725	0	Low
Smith et al. 2021	27	35	Loneliness	0.679	0.093	0.089	0.298	0.026	1	High
Wuthrich & Rapee 2013	27	35	Wellbeing and quality of life	0.511	0.091	0.089	0.298	0.090	0	Some concerns

Mental health data

A general overview of the main data, we use for analyses of mental health outcomes can be found the scroll box below.

mental_health_dat <- 
  gb_dat |> 
  filter(outcome_construct == "Mental health outcome") 

saveRDS(mental_health_dat, file = "mental_health_dat.rds")

mental_overview_dat <- 
  mental_health_dat |> 
  select(
    study, eppi_id, esid, N_t, N_c, N_total, inv_sample_size, gt_pop, vgt_pop, Wgt_pop, gt, vgt, Wgt, Wse, 
    prereg_chr, conventional, analysis_plan, Overall, D5, D7, timing
  )

mental_overview_dat |> 
  mutate(
    p_val = 2 * ( 1 - pnorm( abs(gt_pop) / sqrt(vgt_pop) ) )
  ) |> 
  select(
    `Authors (year)` = study, N_t, N_c,
    `Outcome construct` = analysis_plan, gt_pop, vgt_pop, Wgt, Wse, p_val, 
    `No protocol` = conventional, `Overall RoB` = Overall
  ) |> 
  kable(digits=3)  |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "300px", fixed_thead = TRUE)

Table 2: Data with mental health outcomes.

Authors (year)	N_t	N_c	Outcome construct	gt_pop	vgt_pop	Wgt	Wse	p_val	No protocol	Overall RoB
Acarturk et al. 2022	24	22	General mental health	0.534	0.087	0.084	0.289	0.070	0	Some concerns
Acarturk et al. 2022	24	22	General mental health	0.257	0.093	0.092	0.303	0.398	0	High
Acarturk et al. 2022	24	22	General mental health	0.559	0.135	0.131	0.362	0.128	0	Some concerns
Acarturk et al. 2022	24	22	General mental health	0.339	0.134	0.133	0.364	0.354	0	High
Baekkelund et al. 2022	29	30	General mental health	0.112	0.163	0.163	0.403	0.782	0	Low
Baekkelund et al. 2022	29	30	General mental health	0.100	0.140	0.140	0.375	0.789	0	Low
Cano-Vindel et al. 2021	315	316	Anxiety	0.906	0.004	0.003	0.054	0.000	0	Low
Cano-Vindel et al. 2021	315	316	Depression	0.755	0.005	0.001	0.032	0.000	0	Low
Cano-Vindel et al. 2021	315	316	General mental health	0.611	0.007	0.007	0.083	0.000	0	Low
Cano-Vindel et al. 2021	273	238	Anxiety	0.493	0.004	0.004	0.059	0.000	0	Low
Cano-Vindel et al. 2021	273	238	Depression	0.434	0.002	0.001	0.035	0.000	0	Low
Cano-Vindel et al. 2021	273	238	General mental health	0.467	0.009	0.008	0.091	0.000	0	Low
Cano-Vindel et al. 2021	229	204	Anxiety	0.552	0.005	0.004	0.065	0.000	0	Low
Cano-Vindel et al. 2021	229	204	Depression	0.466	0.003	0.001	0.038	0.000	0	Low
Cano-Vindel et al. 2021	229	204	General mental health	0.459	0.010	0.010	0.099	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Anxiety	0.580	0.005	0.005	0.068	0.000	0	Low
Cano-Vindel et al. 2021	208	180	Depression	0.450	0.003	0.002	0.040	0.000	0	Low
Cano-Vindel et al. 2021	208	180	General mental health	0.551	0.011	0.011	0.105	0.000	0	Low
Craigie & Nathan 2009	157	77	Depression	-0.498	0.025	0.022	0.150	0.002	1	Serious
Craigie & Nathan 2009	157	77	Anxiety	-0.549	0.022	0.021	0.145	0.000	1	Serious
Crawford et al. 2012	121	121	Symptoms of psychosis	0.155	0.040	0.040	0.200	0.439	0	Low
Crawford et al. 2012	119	121	Symptoms of psychosis	0.007	0.040	0.040	0.201	0.974	0	Low
Crawford et al. 2012	121	117	Symptoms of psychosis	0.152	0.040	0.040	0.200	0.446	0	Low
Crawford et al. 2012	117	117	Symptoms of psychosis	0.024	0.041	0.041	0.203	0.905	0	Low
Dyck et al. 2000	21	21	Symptoms of psychosis	0.145	0.005	0.004	0.066	0.032	1	Some concerns
Gestel-Timmermans et al. 2012	134	116	General mental health	0.184	0.014	0.014	0.120	0.127	0	High
Gestel-Timmermans et al. 2012	121	100	General mental health	0.076	0.016	0.016	0.125	0.545	0	High
Gatz et al. 2007	136	177	General mental health	-0.052	0.011	0.011	0.103	0.616	1	Serious
Gatz et al. 2007	136	177	General mental health	-0.216	0.016	0.016	0.126	0.087	1	Serious
Gonzalez & Prihoda 2007	8	9	General mental health	0.596	0.205	0.193	0.439	0.188	1	Serious
Gordon et al. 2018	21	15	General mental health	0.066	0.141	0.141	0.375	0.861	1	Some concerns
Gutman et al. 2019	10	10	General mental health	1.140	0.070	0.034	0.184	0.000	1	Serious
Hagen et al. 2005	15	17	General mental health	0.407	0.169	0.166	0.408	0.323	1	High
Hagen et al. 2005	15	17	Anxiety	0.521	0.171	0.166	0.408	0.208	1	High
Hagen et al. 2005	15	15	Depression	0.355	0.169	0.173	0.416	0.387	1	High
Haslam et al. 2019	66	54	Depression	-0.033	0.000	0.000	0.021	0.109	0	High
Haslam et al. 2019	66	54	General mental health	-0.478	0.025	0.024	0.155	0.002	0	High
Hilden et al. 2021	23	12	General mental health	-0.337	0.142	0.140	0.375	0.371	0	High
Hilden et al. 2021	23	12	Anxiety	-0.202	0.141	0.140	0.375	0.591	0	High
Himle et al. 2014	29	29	General mental health	0.597	0.086	0.083	0.288	0.042	0	Some concerns
Himle et al. 2014	29	29	General mental health	0.385	0.084	0.083	0.288	0.184	0	Some concerns
Himle et al. 2014	29	29	General mental health	0.379	0.084	0.083	0.288	0.191	0	Some concerns
Himle et al. 2014	29	29	Anxiety	0.564	0.086	0.083	0.288	0.054	0	Some concerns
Himle et al. 2014	29	29	General mental health	0.628	0.086	0.083	0.288	0.033	0	Some concerns
Himle et al. 2014	29	29	General mental health	0.729	0.088	0.083	0.288	0.014	0	Some concerns
Himle et al. 2014	29	29	General mental health	0.827	0.089	0.083	0.288	0.006	0	Some concerns
Himle et al. 2014	29	29	Anxiety	0.297	0.084	0.083	0.288	0.304	0	Some concerns
Jacob et al. 2010	19	24	Depression	0.482	0.114	0.116	0.341	0.152	1	Serious
Jacob et al. 2010	19	24	Depression	0.379	0.113	0.116	0.341	0.260	1	Serious
James et al. 2004	29	29	General mental health	0.783	0.088	0.083	0.288	0.008	1	Some concerns
James et al. 2004	31	29	General mental health	0.465	0.081	0.079	0.282	0.103	1	Some concerns
Kanie et al. 2019	32	29	General mental health	0.037	0.078	0.078	0.279	0.893	0	Low
Kanie et al. 2019	32	29	Symptoms of psychosis	-0.041	0.078	0.078	0.279	0.882	0	Low
Lim et al. 2020	18	21	Symptoms of psychosis	0.667	0.143	0.137	0.371	0.078	1	Serious
Lim et al. 2020	18	21	Symptoms of psychosis	-0.084	0.137	0.137	0.371	0.821	1	Serious
Lim et al. 2020	18	21	Symptoms of psychosis	0.695	0.144	0.137	0.371	0.067	1	Serious
Lim et al. 2020	18	21	Symptoms of psychosis	-0.644	0.143	0.137	0.371	0.089	1	Serious
Lim et al. 2020	18	21	Symptoms of psychosis	0.574	0.142	0.137	0.371	0.128	1	Serious
Lloyd-Evans et al. 2020	25	10	Anxiety	0.501	0.162	0.158	0.398	0.213	0	Low
Lloyd-Evans et al. 2020	25	10	Depression	0.456	0.157	0.158	0.398	0.250	0	Low
Madigan et al. 2013	42	22	General mental health	0.000	0.081	0.081	0.285	1.000	1	Some concerns
Madigan et al. 2013	40	20	Symptoms of psychosis	-0.122	0.088	0.088	0.296	0.680	1	Some concerns
Madigan et al. 2013	40	20	Depression	0.065	0.088	0.088	0.296	0.827	1	Some concerns
Madigan et al. 2013	32	17	General mental health	-0.024	0.106	0.106	0.326	0.942	1	Some concerns
Madigan et al. 2013	32	19	Symptoms of psychosis	0.156	0.101	0.100	0.317	0.623	1	Some concerns
Madigan et al. 2013	33	11	Depression	0.022	0.133	0.133	0.365	0.952	1	Some concerns
McCay et al. 2006	26	14	Symptoms of psychosis	0.171	0.115	0.114	0.338	0.613	1	Serious
McCay et al. 2006	26	14	Symptoms of psychosis	0.633	0.169	0.164	0.405	0.124	1	Serious
Michalak et al. 2015	36	35	Depression	0.253	0.068	0.067	0.260	0.332	0	Some concerns
Michalak et al. 2015	36	35	Depression	0.577	0.066	0.067	0.260	0.025	0	Some concerns
Michalak et al. 2015	35	35	Depression	0.830	0.074	0.069	0.262	0.002	0	Some concerns
Michalak et al. 2015	35	35	Depression	0.526	0.067	0.069	0.262	0.042	0	Some concerns
Morley et al. 2014	122	63	General mental health	-0.475	0.028	0.027	0.166	0.005	1	Some concerns
Morley et al. 2014	122	63	Depression	0.021	0.027	0.027	0.166	0.898	1	Some concerns
Morley et al. 2014	122	63	Anxiety	-0.205	0.028	0.027	0.166	0.216	1	Some concerns
Morton et al. 2012	21	20	General mental health	0.758	0.103	0.096	0.309	0.018	1	Some concerns
Morton et al. 2012	21	20	General mental health	0.554	0.052	0.048	0.219	0.015	1	Some concerns
Morton et al. 2012	21	20	General mental health	0.280	0.049	0.048	0.219	0.205	1	Some concerns
Morton et al. 2012	21	20	General mental health	0.481	0.051	0.048	0.219	0.033	1	Some concerns
Patterson et al. 2003	16	16	Symptoms of psychosis	0.451	0.166	0.162	0.403	0.269	1	Some concerns
Patterson et al. 2003	16	16	Symptoms of psychosis	0.962	0.178	0.162	0.403	0.023	1	Some concerns
Patterson et al. 2003	16	16	Symptoms of psychosis	0.449	0.166	0.162	0.403	0.270	1	Some concerns
Patterson et al. 2003	16	16	Depression	0.625	0.169	0.162	0.403	0.128	1	Some concerns
Patterson et al. 2003	16	16	Symptoms of psychosis	0.558	0.168	0.162	0.403	0.173	1	Some concerns
Patterson et al. 2003	16	16	Symptoms of psychosis	0.874	0.176	0.162	0.403	0.037	1	Some concerns
Patterson et al. 2003	16	16	Symptoms of psychosis	0.563	0.168	0.162	0.403	0.170	1	Some concerns
Patterson et al. 2003	16	16	Depression	0.739	0.172	0.162	0.403	0.075	1	Some concerns
Popolo et al. 2019	8	10	General mental health	0.421	0.329	0.324	0.569	0.464	1	Some concerns
Popolo et al. 2019	8	10	General mental health	0.876	0.269	0.245	0.495	0.091	1	Some concerns
Rabenstein et al. 2016	153	148	General mental health	-0.367	0.018	0.018	0.135	0.007	1	Moderate
Rabenstein et al. 2016	153	148	Depression	-0.421	0.017	0.017	0.132	0.001	1	Moderate
Rusch et al. 2019	18	17	Depression	0.648	0.092	0.086	0.293	0.033	0	Some concerns
Rusch et al. 2019	20	13	Depression	0.366	0.183	0.181	0.425	0.392	0	Some concerns
Sacks et al. 2011	38	38	Depression	-0.048	0.066	0.068	0.262	0.852	1	Some concerns
Sacks et al. 2011	38	38	General mental health	-0.064	0.068	0.068	0.262	0.806	1	Some concerns
Sajatovic et al. 2009	63	65	Depression	0.234	0.039	0.039	0.198	0.239	1	Some concerns
Sajatovic et al. 2009	63	65	General mental health	0.302	0.040	0.039	0.198	0.130	1	Some concerns
Sajatovic et al. 2009	51	55	Depression	0.212	0.048	0.048	0.218	0.334	1	Some concerns
Sajatovic et al. 2009	51	55	General mental health	0.104	0.048	0.048	0.218	0.635	1	Some concerns
Sajatovic et al. 2009	41	39	Depression	0.108	0.062	0.062	0.249	0.665	1	Some concerns
Sajatovic et al. 2009	41	39	General mental health	0.189	0.062	0.062	0.249	0.448	1	Some concerns
Saloheimo et al. 2016	27	34	Depression	0.482	0.075	0.073	0.269	0.078	0	Some concerns
Schrank et al. 2016	43	41	General mental health	0.151	0.062	0.061	0.248	0.542	0	Low
Schafer et al. 2019	115	117	Depression	0.081	0.022	0.022	0.150	0.584	0	Some concerns
Schafer et al. 2019	111	117	Depression	0.162	0.022	0.023	0.152	0.280	0	Some concerns
Schafer et al. 2019	115	117	Depression	0.210	0.022	0.022	0.150	0.155	0	Some concerns
Schafer et al. 2019	111	117	Depression	0.186	0.022	0.023	0.152	0.214	0	Some concerns
Schafer et al. 2019	115	117	Depression	0.032	0.022	0.022	0.150	0.827	0	Some concerns
Schafer et al. 2019	111	117	Depression	0.283	0.022	0.023	0.152	0.059	0	Some concerns
Schafer et al. 2019	115	117	General mental health	0.137	0.023	0.022	0.150	0.360	0	Some concerns
Schafer et al. 2019	111	117	General mental health	0.103	0.023	0.023	0.152	0.497	0	Some concerns
Schafer et al. 2019	115	117	General mental health	0.296	0.023	0.022	0.150	0.049	0	Some concerns
Schafer et al. 2019	111	117	General mental health	0.182	0.023	0.023	0.152	0.230	0	Some concerns
Schafer et al. 2019	115	117	General mental health	0.080	0.022	0.022	0.150	0.595	0	Some concerns
Schafer et al. 2019	111	117	General mental health	0.102	0.023	0.023	0.152	0.500	0	Some concerns
Schafer et al. 2019	115	117	General mental health	-0.050	0.022	0.022	0.150	0.739	0	Some concerns
Schafer et al. 2019	111	117	General mental health	-0.089	0.023	0.023	0.152	0.558	0	Some concerns
Schafer et al. 2019	115	117	General mental health	0.190	0.023	0.022	0.150	0.205	0	Some concerns
Schafer et al. 2019	111	117	General mental health	-0.110	0.023	0.023	0.152	0.469	0	Some concerns
Schafer et al. 2019	115	117	General mental health	0.035	0.022	0.022	0.150	0.816	0	Some concerns
Schafer et al. 2019	111	117	General mental health	-0.026	0.023	0.023	0.152	0.866	0	Some concerns
Somers et al. 2017	90	100	General mental health	-0.142	0.034	0.033	0.183	0.437	0	Low
Tjaden et al. 2021	70	67	General mental health	-0.046	0.038	0.038	0.194	0.812	0	Low
Tjaden et al. 2021	63	58	General mental health	-0.061	0.043	0.042	0.206	0.767	0	Low
Valiente et al. 2022	52	61	General mental health	0.046	0.047	0.047	0.218	0.832	0	Low
Valiente et al. 2022	52	61	General mental health	0.021	0.047	0.047	0.218	0.923	0	Low
Valiente et al. 2022	52	61	General mental health	0.000	0.047	0.047	0.218	1.000	0	Low
Valiente et al. 2022	52	61	General mental health	0.153	0.048	0.047	0.218	0.482	0	Low
Valiente et al. 2022	52	61	General mental health	-0.131	0.048	0.047	0.218	0.547	0	Low
Valiente et al. 2022	52	61	General mental health	0.025	0.047	0.047	0.218	0.909	0	Low
Valiente et al. 2022	52	61	General mental health	-0.043	0.047	0.047	0.218	0.844	0	Low
Valiente et al. 2022	52	61	General mental health	0.033	0.047	0.047	0.218	0.881	0	Low
Valiente et al. 2022	52	61	General mental health	0.157	0.048	0.047	0.218	0.470	0	Low
Volpe et al. 2015	21	20	General mental health	0.213	0.127	0.126	0.355	0.549	1	Some concerns
Volpe et al. 2015	21	20	Depression	0.125	0.124	0.126	0.355	0.723	1	Some concerns
Wojtalik et al. 2022	26	23	General mental health	0.084	0.022	0.021	0.147	0.566	0	Low
Wuthrich & Rapee 2013	27	35	Anxiety	1.654	0.113	0.089	0.298	0.000	0	Some concerns
Smith et al. 2021	27	35	Anxiety	1.346	0.105	0.089	0.298	0.000	1	Some concerns
Smith et al. 2021	27	35	Depression	1.801	0.119	0.089	0.298	0.000	1	Some concerns
Wuthrich & Rapee 2013	27	35	Depression	0.811	0.095	0.089	0.298	0.008	0	Some concerns
Wuthrich & Rapee 2013	27	35	Anxiety	0.462	0.091	0.089	0.298	0.125	0	Some concerns

Risk of bias

In this section, we present the main risk og bias (RoB) visualizations.

Note

Note that in plots where the number of studies appears on the x-axis, some bars differ in length. This occurs because a single study contribute multiple effect size estimates that received different RoB assessments.

RoB2

rho <- 0.8

V_mat <- metafor::vcalc(vi = vgt_pop, cluster = study, obs = esid, data = reintegration_dat, rho = rho) 

rma_res <- 
  metafor::rma.mv(
    gt_pop, 
    V = V_mat,
    random = ~ 1 | study / esid,
    data = reintegration_dat
  )

tau2 <- rma_res$sigma2[1]
omega2<- rma_res$sigma2[2]

reint_rob2_dat <- 
  reintegration_dat |> 
  mutate(
    kj = n(),
    sigma2j = mean(vgt_pop),
    weight = 1 / (kj*tau2 + omega2 + (kj-1)*rho*sigma2j + sigma2j),
    .by = study
  ) |> 
  filter(rob_tool == "RoB2") |> 
  select(
    prereg = prereg_chr, study, D1:D5, Overall, weight
  ) 


rob_sum <-  
  reint_rob2_dat |> 
  pivot_longer(D1:Overall, names_to = "Category", values_to = "Rating") |> 
  mutate(
    Category = factor(
      Category, levels = c(paste0("D", 1:5), "Overall"), 
      labels = c(
        "Bias arising from the randomization process",
        "Bias due to deviations from intended interventions",
        "Bias due to missing outcome data",
        "Bias in measurement of the outcome",
        "Bias in selection of the reported result",
        "Overall risk of bias"
        )
      ),
    
    Rating = case_when(str_detect(Rating, "Some") ~ "Some concerns", .default = Rating),
    
    Rating = factor(Rating, levels = c("Low", "Some concerns", "High"))
    
  ) |> 
  group_by(prereg, Category, Rating, study) |> 
  summarize(
    effects = n(),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  summarise(
    studies = length(unique(study)),
    effects = sum(effects),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  mutate(
    effects_pct = 100 * effects / sum(effects),
    studies_pct = 100 * studies / sum(studies),
    effects_pct_w = weight/sum(weight) * 100,
    studies_w = studies * weight,
    studies_pct_w = studies_w / sum(studies_w) * 100
  ) |> 
  ungroup() 

# Mental health
V_mat_mental <- metafor::vcalc(vi = vgt_pop, cluster = study, obs = esid, data = mental_health_dat, rho = rho) 

rma_res_mental <- 
  metafor::rma.mv(
    gt_pop, 
    V = V_mat_mental,
    random = ~ 1 | study / esid,
    data = mental_health_dat
  )

tau2_mental <- rma_res_mental$sigma2[1]
omega2_mental <- rma_res_mental$sigma2[2]

mental_rob2_dat <- 
  mental_health_dat |> 
  mutate(
    kj = n(),
    sigma2j = mean(vgt_pop),
    weight = 1 / (kj*tau2_mental + omega2_mental + (kj-1)*rho*sigma2j + sigma2j),
    .by = study
  ) |> 
  filter(rob_tool == "RoB2") |> 
  select(
    prereg = prereg_chr, study, D1:D5, Overall, weight
  ) 


rob_sum_mental <-  
  mental_rob2_dat |> 
  pivot_longer(D1:Overall, names_to = "Category", values_to = "Rating") |> 
  mutate(
    Category = factor(
      Category, levels = c(paste0("D", 1:5), "Overall"), 
      labels = c(
        "Bias arising from the randomization process",
        "Bias due to deviations from intended interventions",
        "Bias due to missing outcome data",
        "Bias in measurement of the outcome",
        "Bias in selection of the reported result",
        "Overall risk of bias"
        )
      ),
    
    Rating = case_when(str_detect(Rating, "Some") ~ "Some concerns", .default = Rating),
    
    Rating = factor(Rating, levels = c("Low", "Some concerns", "High"))
    
  ) |> 
  group_by(prereg, Category, Rating, study) |> 
  summarize(
    effects = n(),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  summarise(
    studies = length(unique(study)),
    effects = sum(effects),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  mutate(
    effects_pct = 100 * effects / sum(effects),
    studies_pct = 100 * studies / sum(studies),
    effects_pct_w = weight/sum(weight) * 100,
    studies_w = studies * weight,
    studies_pct_w = studies_w / sum(studies_w) * 100
  ) |> 
  ungroup() 

Overall (Not preregistered vs. preregistered)

Reintegration

Number of studies/effects

rob_pct_studies <- 
  rob_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Study-level"
    ) |> 
  ggplot(aes(x = studies, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Number of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "none",
    strip.text.x = element_text(size = 14),
    strip.text.y = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  )  + 
  facetted_pos_scales(
    x = list(
      prereg == "Not preregistered" ~ scale_x_continuous(breaks = seq(0, 16, 2)
      )
    )
  )

rob_pct_effects <- 
  rob_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Outcome-level"
    ) |> 
  ggplot(aes(x = effects, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    x = "Number of effects",
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_blank(),
    strip.text.y = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  ) + 
  facetted_pos_scales(
    x = list(
      prereg == "Preregistered" ~ scale_x_continuous(breaks = seq(0, 150, 25)
      )
    )
  )

#png(filename = "Figures/rob2_reint.png", width = 12, height = 7, units = "in", res = 600)
rob_pct_studies / rob_pct_effects
#dev.off()

Figure 1: RoB2 plot for reintegration studies and outcomes across type of registration (raw numbers)

Weighted percentage

rob_pct_studies_weight <- 
  rob_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "none",
    axis.title.x=element_blank()
  ) 

rob_pct_effects_weight <- 
  rob_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    x = "Weighted percentage",
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_blank()
  ) 


rob_pct_studies_weight / rob_pct_effects_weight

Figure 2: RoB2 plot for reintegration studies and outcomes across type of registration (weighted percentage)

Raw percentage

rob_pct_studies_unweight <- 
  rob_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "none",
    axis.title.x=element_blank()
  ) 

rob_pct_effects_unweight <- 
  rob_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    x = "Percentage",
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_blank()
  ) 


rob_pct_studies_unweight / rob_pct_effects_unweight

Figure 3: RoB2 plot for reintegration studies and outcomes across type of registration (unweighted percentage)

Mental health

Number of studies/effects

rob_pct_studies_mental <- 
  rob_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Study-level"
    ) |> 
  ggplot(aes(x = studies, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Number of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "none",
    strip.text.x = element_text(size = 14),
    strip.text.y = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  )  + 
  facetted_pos_scales(
    x = list(
      prereg == "Not preregistered" ~ scale_x_continuous(breaks = seq(0, 16, 2)
      )
    )
  )

rob_pct_effects_mental <- 
  rob_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Outcome-level"
    ) |> 
  ggplot(aes(x = effects, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    x = "Number of effects",
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_blank(),
    strip.text.y = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  ) 

#png(filename = "Figures/rob2_mental.png", width = 12, height = 7, units = "in", res = 600)
rob_pct_studies_mental / rob_pct_effects_mental
#dev.off()

Figure 4: RoB2 plot for mental health studies and outcomes across type of registration (raw numbers)

Weighted percentage

rob_pct_studies_weight_mental <- 
  rob_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "none",
    axis.title.x=element_blank()
  ) 

rob_pct_effects_weight_mental <- 
  rob_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    x = "Weighted percentage",
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_blank()
  ) 


rob_pct_studies_weight_mental / rob_pct_effects_weight_mental

Figure 5: RoB2 plot for mental health studies and outcomes across type of registration (weighted percentage)

Raw percentage

rob_pct_studies_unweight_mental <- 
  rob_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "none",
    axis.title.x=element_blank()
  ) 

rob_pct_effects_unweight <- 
  rob_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(level ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    x = "Percentage",
    y = "",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_blank()
  ) 


rob_pct_studies_unweight / rob_pct_effects_unweight

Figure 6: RoB2 plot for mental health studies and outcomes across type of registration (unweighted percentage)

Subgrouped risk of bias plots

reint_rob2_subgrp_dat <- 
  reintegration_dat |> 
  mutate(
    kj = n(),
    sigma2j = mean(vgt_pop),
    weight = 1 / (kj*tau2 + omega2 + (kj-1)*rho*sigma2j + sigma2j),
    .by = study
  ) |> 
  filter(rob_tool == "RoB2" & str_detect(analysis_plan, "Alco|Well|Hope|Social")) |> 
  select(
    outcome = analysis_plan, prereg = prereg_chr, study, D1:D5, Overall, weight
  ) |> 
  mutate(
    outcome = case_match(
      outcome,
      "Alcohol and drug abuse/misuse" ~ "Alcohol/drugs",
      "Hope, empowerment & self-efficacy" ~ "Hope/empower",
      "Social functioning (degree of impairment)" ~ "Social function",
      "Wellbeing and quality of life" ~ "Wellbeing/QoL"
    )
  )


rob_sum_subgrp <-  
  reint_rob2_subgrp_dat |> 
  pivot_longer(D1:Overall, names_to = "Category", values_to = "Rating") |> 
  mutate(
    Category = factor(
      Category, levels = c(paste0("D", 1:5), "Overall"), 
      labels = c(
        "Bias arising from the randomization process",
        "Bias due to deviations from intended interventions",
        "Bias due to missing outcome data",
        "Bias in measurement of the outcome",
        "Bias in selection of the reported result",
        "Overall risk of bias"
        )
      ),
    
    Rating = case_when(str_detect(Rating, "Some") ~ "Some concerns", .default = Rating),
    
    Rating = factor(Rating, levels = c("Low", "Some concerns", "High"))
    
  ) |> 
  group_by(prereg, outcome, Category, Rating, study) |> 
  summarize(
    effects = n(),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  summarise(
    studies = length(unique(study)),
    effects = sum(effects),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  mutate(
    effects_pct = 100 * effects / sum(effects),
    studies_pct = 100 * studies / sum(studies),
    effects_pct_w = weight/sum(weight) * 100,
    studies_w = studies * weight,
    studies_pct_w = studies_w / sum(studies_w) * 100
  ) |> 
  ungroup() 

# Mental health
mental_rob2_subgrp_dat <- 
  mental_health_dat |> 
  mutate(
    kj = n(),
    sigma2j = mean(vgt_pop),
    weight = 1 / (kj*tau2 + omega2 + (kj-1)*rho*sigma2j + sigma2j),
    .by = study
  ) |> 
  filter(rob_tool == "RoB2") |> 
  select(
    outcome = analysis_plan, prereg = prereg_chr, study, D1:D5, Overall, weight
  ) |> 
  mutate(
    outcome = factor(
      outcome, 
      levels = c("Anxiety", "Depression", "General mental health", "Symptoms of psychosis")
    )
  )


rob_sum_subgrp_mental <-  
  mental_rob2_subgrp_dat |> 
  pivot_longer(D1:Overall, names_to = "Category", values_to = "Rating") |> 
  mutate(
    Category = factor(
      Category, levels = c(paste0("D", 1:5), "Overall"), 
      labels = c(
        "Bias arising from the randomization process",
        "Bias due to deviations from intended interventions",
        "Bias due to missing outcome data",
        "Bias in measurement of the outcome",
        "Bias in selection of the reported result",
        "Overall risk of bias"
        )
      ),
    
    Rating = case_when(str_detect(Rating, "Some") ~ "Some concerns", .default = Rating),
    
    Rating = factor(Rating, levels = c("Low", "Some concerns", "High"))
    
  ) |> 
  group_by(prereg, outcome, Category, Rating, study) |> 
  summarize(
    effects = n(),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  summarise(
    studies = length(unique(study)),
    effects = sum(effects),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  mutate(
    effects_pct = 100 * effects / sum(effects),
    studies_pct = 100 * studies / sum(studies),
    effects_pct_w = weight/sum(weight) * 100,
    studies_w = studies * weight,
    studies_pct_w = studies_w / sum(studies_w) * 100
  ) |> 
  ungroup() 

Reintegration

Study-level

Number of studies

rob_studies_subgrp <- 
  rob_sum_subgrp |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_nested_wrap(outcome ~ prereg, scales = "free_x", ncol = 2) + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Number of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  )  + 
  facetted_pos_scales(
    x = list(
      prereg == "Not preregistered" & outcome == "Alcohol/drugs" ~ scale_x_continuous(breaks = seq(0, 2, 1)),
      prereg == "Preregistered" & outcome == "Wellbeing/QoL" ~ scale_x_continuous(breaks = seq(0, 14, 2)),
      prereg == "Preregistered" & outcome == "Social function" ~ scale_x_continuous(breaks = seq(0, 10, 2))
    )
  )

rob_studies_subgrp

Figure 7: RoB2 plot for reintegration studies and outcomes by type of registration and outcome (number of studies)

Weighted percentage of studies

# Weigthed percent
rob_weight_pct_studies_subgrp <- 
  rob_sum_subgrp |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_weight_pct_studies_subgrp

Figure 8: RoB2 plot for reintegration studies and outcomes by type of registration and outcome (weighted percentage of studies)

Unweighted percentage of studies

# unweigthed percent
rob_unweight_pct_studies_subgrp <- 
  rob_sum_subgrp |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Percentage of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_unweight_pct_studies_subgrp

Figure 9: RoB2 plot for reintegration studies and outcomes by type of registration and outcome (unweighted percentage of studies)

Outcome-level

Number of effects

rob_effects_subgrp <- 
  rob_sum_subgrp |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_nested_wrap(outcome ~ prereg, scales = "free_x", ncol = 2) + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Number of effects",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  )  + 
  facetted_pos_scales(
    x = list(
      prereg == "Not preregistered" & outcome == "Hope/empower" ~ scale_x_continuous(breaks = seq(0, 12, 2)),
      prereg == "Preregistered" & outcome == "Hope/empower" ~ scale_x_continuous(breaks = seq(0, 20, 5)),
      prereg == "Not preregistered" & outcome == "Social function" ~ scale_x_continuous(breaks = seq(0, 10, 2)),
      prereg == "Preregistered" & outcome == "Social function" ~ scale_x_continuous(breaks = seq(0, 40, 10)),
      prereg == "Not preregistered" & outcome == "Wellbeing/QoL" ~ scale_x_continuous(breaks = seq(0, 10, 2))
    )
  )

rob_effects_subgrp

Figure 10: RoB2 plot for reintegration studies and outcomes by type of registration and outcome (number of effects)

Weighted percentage of effects

# Weigthed percent
rob_weight_pct_effects_subgrp <- 
  rob_sum_subgrp |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage of effects",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_weight_pct_effects_subgrp

Figure 11: RoB2 plot for reintegration studies and outcomes by type of registration and outcome (weighted percentage of effects)

Unweighted percentage of effects

# unweigthed percent
rob_unweight_pct_effects_subgrp <- 
  rob_sum_subgrp |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = effects_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Percentage of effects",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_unweight_pct_effects_subgrp

Figure 12: RoB2 plot for reintegration studies and outcomes by type of registration and outcome (unweighted percentage of effects)

Mental health

Study-level

Number of studies

rob_studies_subgrp_mental <- 
  rob_sum_subgrp_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_nested_wrap(outcome ~ prereg, scales = "free_x", ncol = 2) + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Number of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  )  + 
  facetted_pos_scales(
    x = list(
      prereg == "Not preregistered" & outcome == "Anxiety" ~ scale_x_continuous(breaks = seq(0, 2, 1)),
      prereg == "Preregistered" & outcome == "Depression" ~ scale_x_continuous(breaks = seq(0, 10, 2)),
      prereg == "Not preregistered" & outcome == "General mental health" ~ scale_x_continuous(breaks = seq(0, 10, 2)),
      prereg == "Preregistered" & outcome == "Symptoms of psychosis" ~ scale_x_continuous(breaks = seq(0, 2, 1))
    )
  )

rob_studies_subgrp_mental

Figure 13: RoB2 plot for mental studies and outcomes by type of registration and outcome (number of studies)

Weighted percentage of studies

# Weigthed percent
rob_weight_pct_studies_subgrp_mental <- 
  rob_sum_subgrp_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_weight_pct_studies_subgrp_mental

Figure 14: RoB2 plot for mental health studies and outcomes by type of registration and outcome (weighted percentage of studies)

Unweighted percentage of studies

# unweigthed percent
rob_unweight_pct_studies_subgrp_mental <- 
  rob_sum_subgrp_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Percentage of studies",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_unweight_pct_studies_subgrp_mental

Figure 15: RoB2 plot for mental health studies and outcomes by type of registration and outcome (unweighted percentage of studies)

Outcome-level

Number of effects

rob_effects_subgrp_mental <- 
  rob_sum_subgrp_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_nested_wrap(outcome ~ prereg, scales = "free_x", ncol = 2) + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Number of effects",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  )  + 
  facetted_pos_scales(
    x = list(
      prereg == "Not preregistered" & outcome == "Anxiety" ~ scale_x_continuous(breaks = seq(0, 2, 1)),
      prereg == "Preregistered" & outcome == "Anxiety" ~ scale_x_continuous(breaks = seq(0, 10, 2)),
      prereg == "Not preregistered" & outcome == "Depression" ~ scale_x_continuous(breaks = seq(0, 10, 2)),
      prereg == "Not preregistered" & outcome == "Symptoms of psychosis" ~ scale_x_continuous(breaks = seq(0, 10, 2))
    )
  )

rob_effects_subgrp_mental

Figure 16: RoB2 plot for mental studies and outcomes by type of registration and outcome (number of effects)

Weighted percentage of effects

# Weigthed percent
rob_weight_pct_effects_subgrp_mental <- 
  rob_sum_subgrp_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage of effects",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_weight_pct_effects_subgrp_mental

Figure 17: RoB2 plot for mental health studies and outcomes by type of registration and outcome (weighted percentage of effects)

Unweighted percentage of effects

# unweigthed percent
rob_unweight_pct_effects_subgrp_mental <- 
  rob_sum_subgrp_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = effects_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid2(outcome ~ prereg, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Some concerns" = "lightgoldenrodyellow", "High" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of the outcome",
      "Bias due to missing outcome data",
      "Bias due to deviations from intended interventions",
      "Bias arising from the randomization process"
    )
  ) +
  labs(
    y = "",
    x = "Percentage of effects",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

rob_unweight_pct_effects_subgrp_mental

Figure 18: RoB2 plot for mental health studies and outcomes by type of registration and outcome (unweighted percentage of effects)

ROBINS-I

reint_robinsi_dat <- 
  reintegration_dat |> 
  mutate(
    kj = n(),
    sigma2j = mean(vgt_pop),
    weight = 1 / (kj*tau2 + omega2 + (kj-1)*rho*sigma2j + sigma2j),
    .by = study
  ) |> 
  filter(rob_tool == "ROBINS-I") |> 
  select(
    prereg = prereg_chr, study, D1:Overall, weight
  ) 


robinsi_sum <-  
  reint_robinsi_dat |> 
  pivot_longer(D1:Overall, names_to = "Category", values_to = "Rating") |> 
  mutate(
    Category = factor(
      Category, levels = c(paste0("D", 1:7), "Overall"), 
      labels = c(
        "Bias due to confounding",
        "Bias due to selection of participants",
        "Bias in classification of interventions",
        "Bias due to deviations from intended interventions",
        "Bias due to missing data",
        "Bias in measurement of outcomes",
        "Bias in selection of the reported result",
        "Overall risk of bias"
        )
      ),
    
    Rating = factor(Rating, levels = c("Low", "Moderate", "Serious"))
    
  ) |> 
  group_by(Category, Rating, study) |> 
  summarize(
    effects = n(),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  summarise(
    studies = length(unique(study)),
    effects = sum(effects),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  mutate(
    effects_pct = 100 * effects / sum(effects),
    studies_pct = 100 * studies / sum(studies),
    effects_pct_w = weight/sum(weight) * 100,
    studies_w = studies * weight,
    studies_pct_w = studies_w / sum(studies_w) * 100
  ) |> 
  ungroup() 

# Mental health
mental_robinsi_dat <- 
  mental_health_dat |> 
  mutate(
    kj = n(),
    sigma2j = mean(vgt_pop),
    weight = 1 / (kj*tau2 + omega2 + (kj-1)*rho*sigma2j + sigma2j),
    .by = study
  ) |> 
  filter(rob_tool == "ROBINS-I") |> 
  select(
    prereg = prereg_chr, study, D1:Overall, weight
  ) 


robinsi_sum_mental <-  
  mental_robinsi_dat |> 
  pivot_longer(D1:Overall, names_to = "Category", values_to = "Rating") |> 
  mutate(
    Category = factor(
      Category, levels = c(paste0("D", 1:7), "Overall"), 
      labels = c(
        "Bias due to confounding",
        "Bias due to selection of participants",
        "Bias in classification of interventions",
        "Bias due to deviations from intended interventions",
        "Bias due to missing data",
        "Bias in measurement of outcomes",
        "Bias in selection of the reported result",
        "Overall risk of bias"
        )
      ),
    
    Rating = factor(Rating, levels = c("Low", "Moderate", "Serious"))
    
  ) |> 
  group_by(Category, Rating, study) |> 
  summarize(
    effects = n(),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  summarise(
    studies = length(unique(study)),
    effects = sum(effects),
    weight = sum(weight),
    .groups = "drop_last"
  ) |> 
  mutate(
    effects_pct = 100 * effects / sum(effects),
    studies_pct = 100 * studies / sum(studies),
    effects_pct_w = weight/sum(weight) * 100,
    studies_w = studies * weight,
    studies_pct_w = studies_w / sum(studies_w) * 100
  ) |> 
  ungroup() 

Reintegration

Raw numbers

robin_studies <- 
  robinsi_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Raw number",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.x = element_text(size = 14),
    strip.text.y = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  ) 

robin_effects <- 
  robinsi_sum  |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Raw number",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.ticks.y = element_blank(),
    axis.text.y = element_blank(),
    strip.text.x = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  ) 


#png(filename = "Figures/ROBINSI_reint.png", width = 12, height = 5, units = "in", res = 600)
(robin_studies + robin_effects) +
  plot_layout(guides = "collect", axis_titles = "collect") & theme(legend.position = 'bottom') 
#dev.off()

Figure 19: ROBINS-I plot for reintegration QES studies and outcomes (raw number)

Weighted percentage

robin_pct_studies_weight <- 
  robinsi_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

robin_pct_effects_weight <- 
  robinsi_sum  |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "% Effects"
    ) |> 
  ggplot(aes(x = effects_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.ticks.y = element_blank(),
    axis.text.y = element_blank()
  ) 


(robin_pct_studies_weight + robin_pct_effects_weight) +
  plot_layout(guides = "collect", axis_titles = "collect") & theme(legend.position = 'bottom') 

Figure 20: ROBINS-I plot for reintegration QES studies and outcomes (weighted percentage)

Unweighted percentage

robin_pct_studies_unweight <- 
  robinsi_sum |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

robin_pct_effects_unweight <- 
  robinsi_sum  |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.ticks.y = element_blank(),
    axis.text.y = element_blank()
  ) 




(robin_pct_studies_unweight + robin_pct_effects_unweight) +
  plot_layout(guides = "collect", axis_titles = "collect") & theme(legend.position = 'bottom') 

Figure 21: ROBINS-I plot for reintegration QES studies and outcomes (unweighted percentage)

Mental health

Raw numbers

robin_studies_mental <- 
  robinsi_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Raw number",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    strip.text.y = element_blank(),
    strip.text.x = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  ) 

robin_effects_mental <- 
  robinsi_sum_mental  |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Raw number",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.ticks.y = element_blank(),
    axis.text.y = element_blank(),
    strip.text.x = element_text(size = 14),
    axis.title = element_text(size = 14),
    axis.text = element_text(size = 14),
    legend.text = element_text(size = 14),
    legend.title = element_text(size = 14)
  ) 

#png(filename = "Figures/RONINSI_mental.png", width = 12, height = 5, units = "in", res = 600)
(robin_studies_mental + robin_effects_mental) +
  plot_layout(guides = "collect", axis_titles = "collect") & theme(legend.position = 'bottom') 
#dev.off()

Figure 22: ROBINS-I plot for mental health QES studies and outcomes (raw number)

Weighted percentage

robin_pct_studies_weight_mental <- 
  robinsi_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

robin_pct_effects_weight_mental <- 
  robinsi_sum_mental  |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct_w, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Weighted percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.ticks.y = element_blank(),
    axis.text.y = element_blank()
  ) 


(robin_pct_studies_weight_mental + robin_pct_effects_weight_mental) +
  plot_layout(guides = "collect", axis_titles = "collect") & theme(legend.position = 'bottom') 

Figure 23: ROBINS-I plot for mental health QES studies and outcomes (weighted percentage)

Unweighted percentage

robin_pct_studies_unweight_mental <- 
  robinsi_sum_mental |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Studies"
    ) |> 
  ggplot(aes(x = studies_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom"
  ) 

robin_pct_effects_unweight_mental <- 
  robinsi_sum_mental  |> 
  mutate(
    Rating = fct_rev(Rating),
    level = "Effects"
    ) |> 
  ggplot(aes(x = effects_pct, y = Category, fill = Rating)) + 
  geom_col(alpha = 0.9) +
  facet_grid(~ level, scales = "free_x") + 
  scale_fill_manual(
    values = c("Low" = "mediumaquamarine", "Moderate" = "lightgoldenrodyellow", "Serious" = "lightcoral")
  ) +
  scale_y_discrete(
    limits = rev,
    labels = c( 
      expression(bold("Overall risk of bias")),
      "Bias in selection of the reported result",
      "Bias in measurement of outcomes",
      "Bias due to missing data",
      "Bias due to deviations from intended interventions",
      "Bias in classification of interventions",
      "Bias due to selection of participants",
      "Bias due to confounding"
    )
  ) +
  labs(
    y = "",
    x = "Percentage",
    fill = "Risk of Bias"
  ) + 
  guides(fill = guide_legend(reverse = TRUE)) + 
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.ticks.y = element_blank(),
    axis.text.y = element_blank()
  ) 




(robin_pct_studies_unweight_mental + robin_pct_effects_unweight_mental) +
  plot_layout(guides = "collect", axis_titles = "collect") & theme(legend.position = 'bottom') 

Figure 24: ROBINS-I plot for mental health QES studies and outcomes (unweighted percentage)

Descriptives and Dependence Structures

Timeline

# Figure 1: Number of included studies in the meta-analysis by year
G_timeplot <- 
  group_based_dat |> 
  mutate(
    prereg = if_else(
      str_detect(protocol, regex("yes", ignore_case = TRUE)), 
      "Preregistered", "Not preregistered"
    )
  ) |> 
  reframe(year = unique(year), prereg = unique(prereg), .by =  study)

# DONE
timeline_plot <-  ggplot(G_timeplot, aes(x = year, fill = prereg)) + 
  geom_bar(col = "black", alpha = 1, width = 1) +
  scale_x_continuous(breaks = seq(2000, 2022, 1)) + 
  scale_y_continuous(breaks = seq(0, 6.5, 1), limits = c(0, 7), expand = c(0,0)) + 
  theme_bw() + 
  scale_fill_brewer(palette="Paired")+
  theme(
    axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1),
    #panel.border = element_blank(),
    legend.position = "bottom",
    legend.title = element_blank(),
    panel.grid.minor = element_blank(),
    panel.grid.major.x = element_blank(),
    axis.line = element_line(colour = "black")
  ) +
  labs(
    x = "Year of publication",
    y = "Number of studies",
  )

timeline_plot

Figure 25: Number of studies included in meta-analysis by year.

Number effects across effect size metrics

# Number of studies reporting OR
group_based_dat |> 
  summarise(
    n = n(), 
    .by = c(study, effect_size)
  ) |> 
  summarise(
    N_studies = length(effect_size), 
    N_es = sum(n),
    .by = effect_size
  )

# A tibble: 2 × 3
  effect_size N_studies  N_es
  <chr>           <int> <int>
1 SMD                48   388
2 OR                  1     2

Average pre-posttest correlation

Reintegrational outcome

reintegration_dat |> 
  filter(
    str_detect(
      ppcor_method, regex("Cal", ignore_case = TRUE), 
    ) |
    str_detect(
      ppcor_method, regex("From study", ignore_case = TRUE), 
    ) 
  ) |> 
  select(contains("ppcor")) |> 
  pull(ppcor) |> 
  mean()

[1] 0.6229602

Mental health

mental_health_dat|> 
  filter(
    str_detect(
      ppcor_method, regex("Cal", ignore_case = TRUE), 
    ) |
    str_detect(
      ppcor_method, regex("From study", ignore_case = TRUE), 
    ) 
  ) |> 
  select(contains("ppcor")) |> 
  pull(ppcor) |> 
  mean()

[1] 0.5071045

Number of effect size estimates per study

Overall per study

es_plot_per_study <- 
  gb_dat |>
  arrange(desc(study)) |>
  mutate(study = factor(study, unique(study))) |> 
  ggplot(aes(x = study)) +
  geom_bar(aes(fill = outcome_construct)) +
  scale_y_continuous(breaks = seq(5, 40, by = 5)) + 
  theme_minimal() +
  theme(
    legend.position = "bottom"
  ) +
  coord_flip() +
  labs(
    x = paste0("Study (", n_distinct(gb_dat$study), " studies in total)"), 
    y = "Number of Effect Size Estimates",
    fill = "Outcome construct"
  ) +
  guides(fill = guide_legend(reverse=TRUE))

es_plot_per_study

Figure 26: Distribution of number of effect size estimates per study

Across outcome subgroups per study

reintegration_dat |>
ggplot(aes(y = study, fill = gt_pop >= 0)) + 
geom_bar(data = subset(reintegration_dat, gt_pop >= 0), aes(x = after_stat(count)), stat = "count") + 
geom_bar(data = subset(reintegration_dat, gt_pop < 0), aes(x = -after_stat(count)), stat = "count") + 
theme_minimal() +
scale_x_continuous(labels = abs, breaks = scales::breaks_width(5), limits = c(-10, 30)) +
scale_y_discrete(limits=rev) +
scale_fill_manual(
  values = c(met.brewer("Navajo")[2], met.brewer("Navajo")[4]), 
  labels = c("Negative ES", "Non-negative ES"), 
  name = "Effect Size"
) +
labs(x = "Number of Effect Size Estimates", y = "", title = "Reintegration") +
theme(
    legend.position = "bottom",
    #legend.position.inside = c(0.95, 0.05),
    #legend.justification = c(1, 0),
    legend.background = element_blank(),
    plot.title = element_text(hjust = 0.5) 
  )

Figure 27: Distribution of number of effects per study, by direction of the effects for reintegrational outcomes.

mental_health_dat |>
ggplot(aes(y = study, fill = gt_pop >= 0)) + 
geom_bar(data = subset(mental_health_dat, gt_pop >= 0), aes(x = after_stat(count)), stat = "count") + 
geom_bar(data = subset(mental_health_dat, gt_pop < 0), aes(x = -after_stat(count)), stat = "count") + 
theme_minimal() +
scale_x_continuous(labels = abs, breaks = scales::breaks_width(5), limits = c(-5, 15)) +
scale_y_discrete(limits=rev) +
scale_fill_manual(
  values = c(met.brewer("Navajo")[2], met.brewer("Navajo")[4]), 
  labels = c("Negative ES", "Non-negative ES"), 
  name = "Effect Size"
) +
labs(x = "Number of Effect Size Estimates", y = "", title = "Mental Health") +
theme(
    legend.position = "bottom",
    #legend.position.inside = c(0.95, 0.2),
    #legend.justification = c(1, 0),
    legend.background = element_blank(),
    plot.title = element_text(hjust = 0.5) 
  )

Figure 28: Distribution of number of effects per study, by direction of the effects for mental health outcomes.

Overall across all studies and outcomes

# Multi-arms studies
multi_arm_studies <- 
  gb_dat |> 
  filter(trt_id > 1) |> 
  reframe(study = unique(study))
  

# Multi-time-points studies
follow_up_studies <- 
  gb_dat |>
  summarise(
    time_points = length(unique(time_after_end_intervention_weeks)),
    .by = c(study, trt_id, ctr_id)
  ) |>
  filter(time_points > 1)

# Preregistered studies
prereg_studies <- 
  gb_dat |> 
  summarise(
    prereg_chr = unique(prereg_chr),
    n_es = n(),
    .by = study
  ) |> 
  summarise(
    N_studies = n_distinct(study),
    N_es = sum(n_es),
    .by = prereg_chr
  )

study_sample_sizes <- 
  gb_dat |>
  group_by(study, trt_id, ctr_id) |>
  summarise(
    effects = n(),
    participants = max(N_total)
  ) |>
  summarise(
    effects = sum(effects),
    participants = mean(participants),
    ctl_comparisons = n(),
    ctl_arms = paste(ctr_id, collapse = "; ")
  ) |>
  group_by(study, ctl_arms) |>
  summarise(
    effects = sum(effects),
    participants = mean(participants * (1 + ctl_comparisons * (n() - 1) / 2)),
    trt_comparisons = n(),
    trt_arms = paste(trt_id, collapse = "; "),
    ctl_comparisons = mean(ctl_comparisons)
  ) |>
  summarise(
    effects = sum(effects),
    participants = sum(participants),
    trt_comparisons = mean(trt_comparisons),
    ctl_comparisons = sum(ctl_comparisons)
  )

sample_size_summary <-
  study_sample_sizes |>
  summarise(
    studies = n(),
    studies_multiple_tx = sum(trt_comparisons > 1),
    studies_multiple_ctl = sum(ctl_comparisons > 1),
    n_effects = sum(effects),
    mean_effects = mean(effects),
    min_effects = min(effects),
    median_effects = median(effects),
    max_effects = max(effects),
    participants = round(sum(participants))
  ) |> 
  mutate(prereg = prereg_studies$N_studies[1])

n_studies <- sample_size_summary$studies

kable(
  sample_size_summary,
  col.names = c(
    studies = "Studies",
    studies_multiple_tx = "Multi-treatment studies",
    studies_multiple_ctl = "Multi-control studies",
    n_effects = "Effects",
    mean_effects = "Mean",
    min_effects = "Minimum",
    median_effects = "Median",
    max_effects = "Maximum",
    participants = "Participants",
    prereg = "Preregistered studies"
  ),
  digits = 1
) |>
  kable_styling(bootstrap_options = c("striped", "condensed"),
                full_width = FALSE)

Table 3: Data structure for the all data.

Studies	Multi-treatment studies	Multi-control studies	Effects	Mean	Minimum	Median	Maximum	Participants	Preregistered studies
48	3	0	343	7.1	1	4.5	40	5527	23

gb_dat |> 
  summarise(
    es_count = n(),
    .by = study
  ) |>  
  arrange(es_count) |> 
  ggplot(aes(x = es_count)) +
  geom_histogram(binwidth = 0.5, fill = met.brewer("Navajo")[4]) +
  scale_x_continuous(breaks = seq(0, 40, by = 5)) +
  scale_y_continuous(breaks = seq(0, 10, 2)) +
  theme_minimal() +
  labs(x = "Effect Size Estimates per Study", y = "Number of Studies")

Figure 29: Distribution of number of effect size estimates per study

Across outcome subgroups

# Multi-arms studies
multi_arm_studies_reint <- 
  reintegration_dat |> 
  filter(trt_id > 1) |> 
  reframe(study = unique(study))
  

# Multi-time-points studies
follow_up_studies_reint <- 
  reintegration_dat |>
  summarise(
    time_points = length(unique(time_after_end_intervention_weeks)),
    .by = c(study, trt_id, ctr_id)
  ) |>
  filter(time_points > 1)

# Preregistered studies
prereg_studies_reint <- 
  reintegration_dat |> 
  summarise(
    prereg_chr = unique(prereg_chr),
    n_es = n(),
    .by = study
  ) |> 
  summarise(
    N_studies = n_distinct(study),
    N_es = sum(n_es),
    .by = prereg_chr
  )

study_sample_sizes_reint <- 
  reintegration_dat |>
  group_by(study, trt_id, ctr_id) |>
  summarise(
    effects = n(),
    participants = max(N_total)
  ) |>
  summarise(
    effects = sum(effects),
    participants = mean(participants),
    ctl_comparisons = n(),
    ctl_arms = paste(ctr_id, collapse = "; ")
  ) |>
  group_by(study, ctl_arms) |>
  summarise(
    effects = sum(effects),
    participants = mean(participants * (1 + ctl_comparisons * (n() - 1) / 2)),
    trt_comparisons = n(),
    trt_arms = paste(trt_id, collapse = "; "),
    ctl_comparisons = mean(ctl_comparisons)
  ) |>
  summarise(
    effects = sum(effects),
    participants = sum(participants),
    trt_comparisons = mean(trt_comparisons),
    ctl_comparisons = sum(ctl_comparisons)
  )

sample_size_summary_reint <-
  study_sample_sizes_reint |>
  summarise(
    studies = n(),
    studies_multiple_tx = sum(trt_comparisons > 1),
    studies_multiple_ctl = sum(ctl_comparisons > 1),
    n_effects = sum(effects),
    mean_effects = mean(effects),
    min_effects = min(effects),
    median_effects = median(effects),
    max_effects = max(effects),
    participants = round(sum(participants))
  ) |> 
  mutate(prereg = prereg_studies_reint$N_studies[1])

n_studies_reint <- sample_size_summary_reint$studies

kable(
  sample_size_summary_reint,
  col.names = c(
    studies = "Studies",
    studies_multiple_tx = "Multi-treatment studies",
    studies_multiple_ctl = "Multi-control studies",
    n_effects = "Effects",
    mean_effects = "Mean",
    min_effects = "Minimum",
    median_effects = "Median",
    max_effects = "Maximum",
    participants = "Participants",
    prereg = "Preregistered studies"
  ),
  digits = 1
) |>
  kable_styling(bootstrap_options = c("striped", "condensed"),
                full_width = FALSE)

Table 4: Data structure for the reintegrational data.

Studies	Multi-treatment studies	Multi-control studies	Effects	Mean	Minimum	Median	Maximum	Participants	Preregistered studies
45	3	0	202	4.5	1	3	28	5390	22

# Multi-arms studies
multi_arm_studies_mental <- 
  mental_health_dat |> 
  filter(trt_id > 1) |> 
  reframe(study = unique(study))
  

# Multi-time-points studies
follow_up_studies_mental <- 
  mental_health_dat |>
  summarise(
    time_points = length(unique(time_after_end_intervention_weeks)),
    .by = c(study, trt_id, ctr_id)
  ) |>
  filter(time_points > 1)

# Preregistered studies
prereg_studies_mental <- 
  mental_health_dat |> 
  summarise(
    prereg_chr = unique(prereg_chr),
    n_es = n(),
    .by = study
  ) |> 
  summarise(
    N_studies = n_distinct(study),
    N_es = sum(n_es),
    .by = prereg_chr
  )

study_sample_sizes_mental <- 
  mental_health_dat |>
  group_by(study, trt_id, ctr_id) |>
  summarise(
    effects = n(),
    participants = max(N_total)
  ) |>
  summarise(
    effects = sum(effects),
    participants = mean(participants),
    ctl_comparisons = n(),
    ctl_arms = paste(ctr_id, collapse = "; ")
  ) |>
  group_by(study, ctl_arms) |>
  summarise(
    effects = sum(effects),
    participants = mean(participants * (1 + ctl_comparisons * (n() - 1) / 2)),
    trt_comparisons = n(),
    trt_arms = paste(trt_id, collapse = "; "),
    ctl_comparisons = mean(ctl_comparisons)
  ) |>
  summarise(
    effects = sum(effects),
    participants = sum(participants),
    trt_comparisons = mean(trt_comparisons),
    ctl_comparisons = sum(ctl_comparisons)
  )

sample_size_summary_mental <-
  study_sample_sizes_mental |>
  summarise(
    studies = n(),
    studies_multiple_tx = sum(trt_comparisons > 1),
    studies_multiple_ctl = sum(ctl_comparisons > 1),
    n_effects = sum(effects),
    mean_effects = mean(effects),
    min_effects = min(effects),
    median_effects = median(effects),
    max_effects = max(effects),
    participants = round(sum(participants))
  ) |> 
  mutate(prereg = prereg_studies_mental$N_studies[1])

n_studies_mental <- sample_size_summary_mental$studies

kable(
  sample_size_summary_mental,
  col.names = c(
    studies = "Studies",
    studies_multiple_tx = "Multi-treatment studies",
    studies_multiple_ctl = "Multi-control studies",
    n_effects = "Effects",
    mean_effects = "Mean",
    min_effects = "Minimum",
    median_effects = "Median",
    max_effects = "Maximum",
    participants = "Participants",
    prereg = "Preregistered studies"
  ),
  digits = 1
) |>
  kable_styling(bootstrap_options = c("striped", "condensed"),
                full_width = FALSE)

Table 5: Data structure for the mental health data.

Studies	Multi-treatment studies	Multi-control studies	Effects	Mean	Minimum	Median	Maximum	Participants	Preregistered studies
41	3	0	141	3.4	1	2	18	4663	20

reintegration_dat |> 
  summarise(
    es_count = n(),
    .by = study
  ) |>  
  arrange(es_count) |> 
  ggplot(aes(x = es_count)) +
  geom_histogram(binwidth = 0.5, fill = "cornflowerblue") +
  scale_x_continuous(breaks = seq(0, 30, by = 5)) +
  scale_y_continuous(breaks = seq(0, 14, 2), limits = c(0, 14)) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5)) +
  labs(x = "Effect Size Estimates per Study", y = "Number of Studies", title = "Reintegration") +
  expand_limits(x = 30)

Figure 30: Distribution of number of effects per study for mental for reintegrational outcomes.

mental_health_dat |> 
  summarise(
    es_count = n(),
    .by = study
  ) |>  
  arrange(es_count) |> 
  ggplot(aes(x = es_count)) +
  geom_histogram(binwidth = 0.5, fill = "gray") +
  scale_x_continuous(breaks = seq(0, 20, by = 5)) +
  scale_y_continuous(breaks = seq(0, 10, 2), limits = c(0, 11)) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5)) +
  labs(x = "Effect Size Estimates per Study", y = "Number of Studies", title = "Mental Health") +
  expand_limits(x = 20)

Figure 31: Distribution of number of effects per study for mental health outcomes.

Data structure by outcome constructs

label_cat_hist_ridge <- function(data, n_es, variable, label_map, level_order) {
  require(dplyr)
  require(rlang)
  require(tidyr)
  require(ggplot2)
  require(forcats)
  
  n_es_exp <- enquo(n_es)
  var_exp <- enquo(variable)
  
  data <- data |> 
    mutate(
      !!var_exp := fct_recode(!!var_exp, !!!label_map),
      !!var_exp := fct_relevel(!!var_exp, !!!level_order),
      !!var_exp := fct_rev(!!var_exp)
    ) 
  
  ggplot(data, aes(
    x = !!n_es_exp,
    y = !!var_exp,
    fill = !!var_exp
  )) +
    geom_density_ridges(
      alpha = 1,
      stat = "binline",
      bins = 30,
      scale = 0.7,
      linewidth = 0.1,
      draw_baseline = FALSE
    ) +
    theme_minimal() +
    labs(y = "", x = "Effect Size Estimates per Study") +
    theme(legend.position = "none")
}

The following plot shows the effect size estimates distribution within each outcome construct.

n_es_by_construct <- 
  gb_dat |> 
  filter(str_detect(outcome_construct, "Re")) |> 
  summarise(effects = n(), .by = c(study, analysis_plan)) |> 
  mutate(J = n(), .by = c(analysis_plan)) |> 
  arrange(desc(J)) 

# Used to construct out_label 
#label_dat <- 
#  n_es_by_construct |> 
#  summarise(J = n(), N_es = sum(effects), .by = analysis_plan)


out_label <- c(
  "Wellbeing and quality of life (J = 24, K = 67)" = "Wellbeing and quality of life",
  "Social functioning (degree of impairment) (J =16, K = 47)" = "Social functioning (degree of impairment)",
  "Hope, empowerment & self-efficacy (J = 12, K = 32)" = "Hope, empowerment & self-efficacy",
  "Alcohol and drug abuse/misuse (J = 7, K = 31)" = "Alcohol and drug abuse/misuse",
  "Self-esteem (J = 5, K = 14)" = "Self-esteem",
  "Loneliness (J = 4, K = 5)" = "Loneliness",
  "Physical health (J = 2, K = 3)" = "Physical health",
  "Psychiatric hospitalization (J = 1, K = 1)" = "Psychiatric hospitalization",
  "Employment (J = 1, K = 2)" = "Employment"
)

out_level <- c(
 "Wellbeing and quality of life (J = 24, K = 67)",
  "Social functioning (degree of impairment) (J =16, K = 47)",
  "Hope, empowerment & self-efficacy (J = 12, K = 32)",
  "Alcohol and drug abuse/misuse (J = 7, K = 31)",
  "Self-esteem (J = 5, K = 14)",
  "Loneliness (J = 4, K = 5)",
  "Physical health (J = 2, K = 3)",
  "Psychiatric hospitalization (J = 1, K = 1)",
  "Employment (J = 1, K = 2)"
)

label_cat_hist_ridge(
  data = n_es_by_construct, 
  n_es = effects, 
  variable = analysis_plan, 
  label_map = out_label, 
  level_order = out_level
) + 
  expand_limits(x = 20) + 
  labs(title = "Reintegrational outcomes") + 
  theme(plot.title = element_text(hjust = 0.5))

Figure 32: Distribution of effect size estimates, by reintegrational outcome constructs.

n_es_by_construct_mental <- 
  gb_dat |> 
  filter(str_detect(outcome_construct, "Men")) |> 
  summarise(effects = n(), .by = c(study, analysis_plan)) |> 
  mutate(J = n(), .by = c(analysis_plan)) |> 
  arrange(desc(J)) 

# Used to construct out_label_mental 
#label_dat_mental <- 
#  n_es_by_construct_mental |> 
#  summarise(J = n(), N_es = sum(effects), .by = analysis_plan)

out_label_mental <- c(
  "General mental health (J = 28, K = 70)" = "General mental health",
  "Depression (J = 19, K = 36)" = "Depression",
  "Anxiety (J = 9, K = 14)" = "Anxiety",
  "Symptoms of psychosis (J = 7, K = 21)" = "Symptoms of psychosis"
)

out_level_mental <- c(
  "General mental health (J = 28, K = 70)",
  "Depression (J = 19, K = 36)",
  "Anxiety (J = 9, K = 14)",
  "Symptoms of psychosis (J = 7, K = 21)"
)

label_cat_hist_ridge(
  data = n_es_by_construct_mental, 
  n_es = effects, 
  variable = analysis_plan, 
  label_map = out_label_mental, 
  level_order = out_level_mental
) + 
  expand_limits(x = 15) + 
  labs(title = "Mental heath outcomes") + 
  theme(plot.title = element_text(hjust = 0.5))

Figure 33: Distribution of effect size estimates, by mental health outcome constructs.

Primary study sample size

Total sample sizes

# Used when describing the treatment group sample sizes
N_t_stud_trtid <- 
  reintegration_dat |> 
  reframe(N_treat = max(N_t), .by = c(study, trt_id)) 

N_t_total <- 
  N_t_stud_trtid |> 
  summarise(N_t_total = sum(N_treat), .by = study)

N_c_total <- 
  reintegration_dat |> 
  summarise(N_c_total = max(N_c), .by = study)

N_total_dat <- 
  left_join(N_t_total, N_c_total, by = join_by(study)) |> 
  mutate(N_total = N_t_total + N_c_total)


primary_sample_size_descriptive <- 
  N_total_dat$N_total |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) 

primary_sample_size_descriptive |>
  knitr::kable(
    digits = 2,
    booktabs = TRUE
  ) |>
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Table 6: Distribution of primary study sample sizes at post test for reintegrational outcomes

mean	sd	p0	p25	p50	p75	p100
119.78	127.69	17	41	61	137	631

# Used when describing the treatment group sample sizes
N_t_stud_trtid_mental <- 
  mental_health_dat |> 
  reframe(N_treat = max(N_t), .by = c(study, trt_id)) 

N_t_total_mental <- 
  N_t_stud_trtid_mental |> 
  summarise(N_t_total = sum(N_treat), .by = study)

N_c_total_mental <- 
  mental_health_dat |> 
  summarise(N_c_total = max(N_c), .by = study)

N_total_dat_mental <- 
  left_join(N_t_total_mental, N_c_total_mental, by = join_by(study)) |> 
  mutate(N_total = N_t_total + N_c_total)


primary_sample_size_descriptive_mental <- 
  N_total_dat_mental$N_total |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) 

primary_sample_size_descriptive_mental |>
  knitr::kable(
    digits = 2,
    booktabs = TRUE
  ) |>
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Table 7: Distribution of primary study sample sizes at post test for mental health outcomes

mean	sd	p0	p25	p50	p75	p100
113.73	125.6	17	40	61	128	631

The following plot displays the distribution of study sample sizes at post-test.

study_sizes_plot_reint <- 
  ggplot(N_total_dat, aes(N_total)) + 
  geom_density(fill = "cornflowerblue", alpha = 0.8) + 
  geom_blank(aes(x = 0, y = 0)) + 
  geom_rug(alpha = 0.8) +
  scale_y_continuous(NULL, breaks = NULL) + 
  theme_minimal() + 
  labs(x = "Total Sample Size", y = "")
study_sizes_plot_reint

Figure 34: Distribution of primary study sample sizes at post test for reintegrational outcomes.

study_sizes_plot_mental <- 
  ggplot(N_total_dat_mental, aes(N_total)) + 
  geom_density(fill = "gray", alpha = 0.8) + 
  geom_blank(aes(x = 0, y = 0)) + 
  geom_rug(alpha = 0.8) +
  scale_y_continuous(NULL, breaks = NULL) + 
  theme_minimal() + 
  labs(x = "Total Sample Size", y = "")
study_sizes_plot_mental

Figure 35: Distribution of primary study sample sizes at post test for mental health outcomes.

Treatment group

treatment_sample_size_descriptive <- 
  N_t_stud_trtid$N_treat |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) 

treatment_sample_size_descriptive |>
  knitr::kable(
    digits = 2,
    booktabs = TRUE
  ) |>
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Table 8: Distribution of treatment sample sizes at post test for reintegrational outcomes

mean	sd	p0	p25	p50	p75	p100
60.88	60.32	8	22.75	33.5	90.25	315

treatment_sample_size_descriptive_mental <- 
  N_t_stud_trtid_mental$N_treat |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) 

treatment_sample_size_descriptive_mental |>
  knitr::kable(
    digits = 2,
    booktabs = TRUE
  ) |>
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Table 9: Distribution of treatment sample sizes at post test for mental health outcomes

mean	sd	p0	p25	p50	p75	p100
57.8	59.09	8	21	30	75	315

The following plot displays the distribution of treatment sample sizes at post-test.

treatment_sizes_plot_reint <- 
  ggplot(N_t_stud_trtid, aes(N_treat)) + 
  geom_density(fill = "cornflowerblue", alpha = 0.8) + 
  geom_blank(aes(x = 0, y = 0)) + 
  geom_rug(alpha = 0.8) +
  scale_y_continuous(NULL, breaks = NULL) + 
  theme_minimal() + 
  labs(x = "Total Treatment Sample Size", y = "")
treatment_sizes_plot_reint

Figure 36: Distribution of treatment sample sizes at post test for reintegrational outcomes.

treatment_sizes_plot_mental <- 
  ggplot(N_t_stud_trtid_mental, aes(N_treat)) + 
  geom_density(fill = "gray", alpha = 0.8) + 
  geom_blank(aes(x = 0, y = 0)) + 
  geom_rug(alpha = 0.8) +
  scale_y_continuous(NULL, breaks = NULL) + 
  theme_minimal() + 
  labs(x = "Total Treatment Sample Size", y = "")
treatment_sizes_plot_mental

Figure 37: Distribution of treatment sample sizes at post test for mental health outcomes.

Control group

control_sample_size_descriptive <- 
  N_total_dat$N_c_total |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) 

control_sample_size_descriptive |>
  knitr::kable(
    digits = 2,
    booktabs = TRUE
  ) |>
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Table 10: Distribution of control group sample sizes at post test for reintegrational outcomes

mean	sd	p0	p25	p50	p75	p100
54.84	60.5	9	19	30	63	316

control_sample_size_descriptive_mental <- 
  N_total_dat_mental$N_c_total |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) 

control_sample_size_descriptive_mental |>
  knitr::kable(
    digits = 2,
    booktabs = TRUE
  ) |>
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Table 11: Distribution of control group sample sizes at post test for mental health outcomes

mean	sd	p0	p25	p50	p75	p100
51.71	58.61	9	20	29	63	316

The following plot displays the distribution of control sample sizes at post-test.

control_sizes_plot_reint <- 
  ggplot(N_total_dat, aes(N_c_total)) + 
  geom_density(fill = "cornflowerblue", alpha = 0.8) + 
  geom_blank(aes(x = 0, y = 0)) + 
  geom_rug(alpha = 0.8) +
  scale_y_continuous(NULL, breaks = NULL) + 
  theme_minimal() + 
  labs(x = "Total Control Group Sample Size", y = "")
control_sizes_plot_reint

Figure 38: Distribution of control group sample sizes at post test for reintegrational outcomes.

control_sizes_plot_mental <- 
  ggplot(N_total_dat_mental, aes(N_c_total)) + 
  geom_density(fill = "gray", alpha = 0.8) + 
  geom_blank(aes(x = 0, y = 0)) + 
  geom_rug(alpha = 0.8) +
  scale_y_continuous(NULL, breaks = NULL) + 
  theme_minimal() + 
  labs(x = "Total Control Group Sample Size", y = "")
control_sizes_plot_mental

Figure 39: Distribution of control group sample sizes at post test for mental health outcomes.

Study sample sizes versus number of effect size estimates per study

plot <- 
  reintegration_dat |> 
  summarise(effects = n(), .by = study) |> 
  left_join(N_total_dat, by = join_by(study)) |> 
  relocate(effects, .after = N_total) |> 
  ggplot(aes(x = N_total, y = effects)) +
  geom_point(alpha = 0.5) +
  scale_x_continuous(breaks = seq(0, 650, by = 50)) +
  guides(size = "none") + 
  theme_minimal() +
  theme(
    legend.position = "inside",
    legend.position.inside = c(0.9, 0.05),
    legend.justification = c(1, 0),
    legend.background = element_blank()
  ) +
  labs(x = "Total sample size", y = "Number of effect size estimates per study", color = "") +
  expand_limits(y = 30)

ggMarginal(plot, type = "density")

Figure 40: Study sample sizes versus number of effect size estimates for reintegrational outcomes

Moderators

cat_dat_cross <- function(variable, study_id, data) {
  
  require(dplyr)
  require(rlang)
  require(tidyr)
  
  var_exp <- enquo(variable)
  study_id_exp <- enquo(study_id)
  study_id_name <- as_name(study_id_exp)
  
  
    res_dat <- data %>%
      group_by(!!study_id_exp) %>%
      reframe(var_exp_mirror = unique(!!var_exp))  %>%
      full_join(data,
                by = c(study_id_name),
                relationship = "many-to-many") %>%
      group_by(!!var_exp, var_exp_mirror) %>%
      reframe(
        m = n_distinct(!!study_id_exp),
        k = n()
      ) %>%
      mutate(size = paste0(m, " (", k, ")")) %>%
      select(-m, -k) |>
      pivot_wider(names_from = var_exp_mirror, values_from = "size")
  
  names(res_dat)[1] <- "level" 
  
  res_dat <- 
    res_dat |> 
    mutate(
      level = factor(level, levels = unique(names(res_dat[-1])))
    ) |> 
    arrange(level)

  return(res_dat)
}

Categorical moderators

Outcome measure

outcome_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = analysis_plan,
  study_id = study
)


outcome_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Outcome",
      colnames(outcome_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.",
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 12: Dependency table for preregistration status (reintegration)

Outcome	Alcohol and drug abuse/misuse	Hope, empowerment & self-efficacy	Physical health	Social functioning (degree of impairment)	Wellbeing and quality of life	Employment	Self-esteem	Loneliness	Psychiatric hospitalization
Alcohol and drug abuse/misuse	7 (31)	1 (2)	1 (1)	3 (4)	2 (3)	-	-	-	-
Hope, empowerment & self-efficacy	1 (1)	12 (32)	-	1 (2)	6 (14)	1 (6)	3 (15)	1 (4)	1 (1)
Physical health	1 (1)	-	2 (3)	2 (3)	-	-	-	-	-
Social functioning (degree of impairment)	3 (7)	1 (2)	2 (5)	16 (47)	7 (30)	-	-	-	-
Wellbeing and quality of life	2 (3)	6 (12)	-	7 (34)	24 (67)	-	2 (2)	2 (3)	-
Employment	-	1 (2)	-	-	-	1 (2)	1 (2)	-	-
Self-esteem	-	3 (5)	-	-	2 (2)	1 (2)	5 (14)	-	-
Loneliness	-	1 (2)	-	-	2 (3)	-	-	4 (5)	-
Psychiatric hospitalization	-	1 (1)	-	-	-	-	-	-	1 (1)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

outcome_subgroup_dat_cross <- cat_dat_cross(
  data = filter(reintegration_dat, str_detect(analysis_plan, "Alco|Hope|Social|Well")),
  variable = analysis_plan,
  study_id = study
)


outcome_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Outcome",
      colnames(outcome_subgroup_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 13: Dependency table for outcome used for subgroup analysis (reintegration)

Outcome	Alcohol and drug abuse/misuse	Hope, empowerment & self-efficacy	Social functioning (degree of impairment)	Wellbeing and quality of life
Alcohol and drug abuse/misuse	7 (31)	1 (2)	3 (4)	2 (3)
Hope, empowerment & self-efficacy	1 (1)	12 (32)	1 (2)	6 (14)
Social functioning (degree of impairment)	3 (7)	1 (2)	16 (47)	7 (30)
Wellbeing and quality of life	2 (3)	6 (12)	7 (34)	24 (67)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

outcome_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = analysis_plan,
  study_id = study
)


outcome_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Outcome",
      colnames(outcome_subgroup_dat_cross_mental)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 14: Dependency table for outcome used for subgroup analysis (mental health)

Outcome	Anxiety	Depression	General mental health	Symptoms of psychosis
Anxiety	9 (14)	7 (11)	5 (9)	-
Depression	7 (10)	19 (36)	10 (21)	2 (4)
General mental health	5 (13)	10 (27)	28 (70)	2 (3)
Symptoms of psychosis	-	2 (8)	2 (3)	7 (21)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

Ridge plot of effect size estimates

cat_ridge <- function(data, es, v, variable) {
  require(dplyr)
  require(rlang)
  require(tidyr)
  require(ggplot2)
  
  es_exp <- enquo(es)
  var_exp <- enquo(variable)
  v_exp <- enquo(v)
  
  data |> 
    mutate(!!var_exp := fct_rev(!!var_exp)) |> 
  ggplot(aes(
    x = !!es_exp,
    y = !!var_exp,
    fill = !!var_exp
  )) +
    geom_density_ridges(
      aes(
        point_colour = !!var_exp,
        point_size = 1 / !!v_exp
      ),
      alpha = .2,
      point_alpha = 0.5,
      jittered_points = TRUE
    ) +
    theme_minimal() +
    labs(y = "", x = "Standardized Mean Difference (Hedges' g)") +
    theme(legend.position = "none")
}

Subgroup analyzed

analyzed_outcomes <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Alco|Hope|Social|Well"))

cat_ridge(data = analyzed_outcomes, es = gt_pop, variable = analysis_plan, v = vgt_pop)

Figure 41: Distribution of effect size estimates, by reintegrational outcomes

Not subgroup analyzed

not_analyzed_outcomes <- 
  reintegration_dat |> 
  filter(!str_detect(analysis_plan, "Alco|Hope|Social|Well"))

cat_ridge(data = not_analyzed_outcomes, es = gt_pop, variable = analysis_plan, v = vgt_pop)

Figure 42: Distribution of effect size estimates, by reintegrational outcomes.

cat_ridge(data = mental_health_dat, es = gt_pop, variable = analysis_plan, v = vgt_pop)

Figure 43: Distribution of effect size estimates, by mental health outcomes.

Network plot for outcome construct

The following plot shows the network structure of outcomes constructs. Each node represent an outcome construct, the edge between a pair of nodes indicates that there is at least one study that examined the contracts between that pair of constructs The width of the edges indicates the number of studies that compare that pair of constructs. The size of the node corresponds to the number of studies measure that construct.

gb_dat_reduced <- gb_dat |> select(study, analysis_plan)

res_dat <- gb_dat_reduced |> 
  group_by(study) |>
  reframe(var_exp_mirror = unique(analysis_plan))  |>
  full_join(gb_dat_reduced ,
            by = join_by(study),
            relationship = "many-to-many") |>
  group_by(analysis_plan, var_exp_mirror) |>
  reframe(
    m = n_distinct(study),
    k = n()
  ) |>
  mutate(size = m) |> 
  select(-m, -k) |>
  arrange(var_exp_mirror) |> 
  pivot_wider(names_from = var_exp_mirror, values_from = "size") |> 
  arrange(analysis_plan) |> 
  as.data.frame()


names(res_dat)[1] <- "level"


edges <- res_dat  |>
  rename(from = "level") |> 
  pivot_longer(-from, names_to = "to", values_to = "weight") |>
  filter(from != to, !is.na(weight)) |> 
  mutate(
    from_chr = as.character(from),
    to_chr = as.character(to),
    grp_str = paste0(pmin(from_chr, to_chr, na.rm = TRUE), "_", pmax(from_chr, to_chr, na.rm = TRUE))
  ) |> 
  distinct(grp_str, .keep_all = TRUE) |>
  select(from, to, weight)

g <- graph_from_data_frame(edges, directed = FALSE)
layout <- layout_in_circle(g)

# Adjust label position outward
label_coords <- layout
label_coords[, 1] <- c(
  label_coords[1, 1] * 1.35, #1.5,
  label_coords[2:6, 1] * 1.5, #1.63,
  label_coords[7, 1] * 1.35, #1.68,
  label_coords[8:13, 1] * 1.4 #1.63
)

label_coords[, 2] <- c(
  label_coords[c(1:3), 2],
  label_coords[4, 2] * 1.13,
  label_coords[5, 2] * 1.3,
  label_coords[6, 2] * 1.13,
  label_coords[c(7:9), 2] * 1.13,
  label_coords[10, 2] * 1.3,
  label_coords[11, 2] * 1.3,
  label_coords[12, 2] * 1.3,
  label_coords[13, 2] * 1.13
)

node_sizes <- diag(as.matrix(res_dat[, -1]))

plot(
  g,
  layout = layout,
  edge.width = E(g)$weight,
  edge.color = met.brewer("Navajo")[5],
  vertex.size = node_sizes[c(1:8, 10:13, 9)],
  vertex.label = NA,
  vertex.color = met.brewer("Navajo")[4],
  vertex.frame.width = 0
)

text(
  x = label_coords[, 1],
  y = label_coords[, 2],
  labels = paste0(str_wrap(names(res_dat[c(2:9, 11:14, 10)]), width = 15), "\n(J = ", node_sizes, ")"),
  cex = 0.6,
  col = "black",
  xpd = NA
)

Figure 44: Network structure of contrasts between primary and secondary outcome constructs

Diagnosis (schizophrenia vs. rest of the effects)

diagnosis_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = schizophrenia,
  study_id = study
)

diagnosis_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Diagnosis",
      colnames(diagnosis_subgroup_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 15: Dependency table for diagnosis (reintegration)

Diagnosis	Other	Schizophrenia
Other	37 (172)	-
Schizophrenia	-	8 (30)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

diagnosis_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = schizophrenia,
  study_id = study
)

diagnosis_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Diagnosis",
      colnames(diagnosis_subgroup_dat_cross_mental)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 16: Dependency table for diagnosis (mental health)

Diagnosis	Other	Schizophrenia
Other	35 (130)	-
Schizophrenia	-	6 (11)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = schizophrenia, v = vgt_pop)

Figure 45: Distribution of effect size estimates, by diagnosis (reintegrational outcome).

cat_ridge(data = mental_health_dat, es = gt_pop, variable = schizophrenia, v = vgt_pop)

Figure 46: Distribution of effect size estimates, by diagnosis (mental health).

Type of intervention

cbt_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = CBT_int,
  study_id = study
)

cbt_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Intervention",
      colnames(cbt_subgroup_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 17: Dependency table for type of intervention (reintegration)

Intervention	CBT	Other
CBT	11 (52)	1 (5)
Other	1 (5)	35 (150)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cbt_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = CBT_int,
  study_id = study
)

cbt_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Intervention",
      colnames(cbt_subgroup_dat_cross_mental)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 18: Dependency table for type of intervention (mental health)

Intervention	CBT	Other
CBT	11 (46)	1 (2)
Other	1 (2)	31 (95)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = CBT_int, v = vgt_pop)

Figure 47: Distribution of effect size estimates, by type of intervention (reintegrational outcome).

cat_ridge(data = mental_health_dat, es = gt_pop, variable = CBT_int, v = vgt_pop)

Figure 48: Distribution of effect size estimates, by type of intervention (mental health).

EGM-like plot

Reintegrational outcomes

egm_dat_reint <- 
  reintegration_dat |> 
  summarise(
    n = n(),
    n_studies = factor(n_distinct(study), levels = c(1:4)),
    .by = c(trt_group, analysis_plan, prereg_chr)
  ) |> 
  mutate(
    n_prereg = case_match(
      prereg_chr,
      "Preregistered" ~ 0.5,
      "Not preregistered" ~ 0
    ),
    
    constant = "Intervention",
    
    outcome = case_match(
      analysis_plan,
      "Alcohol and drug abuse/misuse" ~ "Alcohol/drugs",
      "Hope, empowerment & self-efficacy" ~ "Hope",
      "Psychiatric hospitalization" ~ "Psych hospital",
      "Social functioning (degree of impairment)" ~ "Social func",
      "Wellbeing and quality of life" ~ "Wellbeing/QoL",
      .default = analysis_plan
    ),
    
    intervention = case_match(
      trt_group,
      "group-based CBT" ~ "GrpCBT",
      "stress management" ~ "StressMan",
      "vocational training" ~ "JobTrain",
      "group psychotherapy"  ~ "GrpPsychT",
      "art therapy" ~ "ArtT",
      "illness management" ~ "IllMan",
      "group psychoeducation" ~ "GrpPsychEdu",
      "seeking safety" ~ "SeekSafe",
      "social cognition and interaction training" ~ "SCIT",
      "illness and addiction management" ~ "Ill&AddMan",
      "social network training" ~ "NetworkTrain",
      "cognitive-behavioral social skills training" ~ "CBSST",
      "residential treatment" ~ "ResidentTreat",
      "positive psychology group intervention" ~ "PosiPsychGrp",
      "addiction management" ~ "AddMan",
      "reading group intervention" ~ "ReadGrp"
    )
    
    
    #intervention_name = case_match(
    #  intervention,
    #  "Sanctions and economic incentives" ~ "Econ incentives",
    #  "Introduction programmes" ~ "Intro program",
    #  "Combination programmes" ~ "Combi program",
    #  "Language training" ~ "Lang train",
    #  "Labour market training" ~ "Lab market train",
    #  "Job search assistance" ~" Job search ass",
    #  "Subsidized public sector employment" ~ "Sub public emp"
    #),
    #
    #outcome = if_else(outcome == "Duration of social assistance spells", "Dur. social ass spells", outcome)
    
  ) 

egm_dat_reint |> 
  ggplot(aes(x = n_prereg, y =  constant, size = n, color = n_studies)) + 
  geom_point() +
  scale_size(range = c(6, 14)) +
  geom_text(aes(label = n), color = "black", size = 3.5) +
  scale_x_continuous(breaks = seq(0,0.5,0.5), limits = c(0, 1), labels = c("Conventional", "Preregistered")) +
  scale_y_discrete("Interventions") +
  facet_grid(intervention~outcome, scales = "free_y", space = "free_y") +
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.text.x = element_text(angle = 45, hjust = 1),
    #panel.grid.major = element_blank(),
    #panel.grid.minor = element_blank(),
    #panel.background = element_blank(),
    axis.text.y = element_blank(),   
    axis.ticks.y = element_blank(),  
    axis.title.y = element_blank(),
    plot.caption = element_text(hjust = 0, size = 14)
  ) +
  coord_cartesian(xlim = c(-0.25, 0.75), ylim = NULL) +
  guides(size = "none") + 
  xlab("Type of registration") +
  labs(
    color = "Number of studies",
    caption = paste0(
      "Note: GrpCBT = Group-based CBT, StressMan = Stress mangement, JobTrain = Vocational training, ",
      "GrpPsychT = Group psychotherapy, ArtT = Art Therapy, IllMan = Ill management,\n",
      "GrpPsychEdu = Group Psychoeducation, SeekSafe = Seeking safety, ",
      "SCIT = cognitive-behavioral social skills training, ResidentTreat = Residential treatment,\n",
      "PosiPsychGrp = Positive Psychology group therapy, AddMan = Addiction mangement, ",
      "ReadGrp = Reading group."
    )
  )

Figure 49: EGM for reintegrational outcomes

Mental health outcomes

egm_dat_mental <- 
  mental_health_dat |> 
  summarise(
    n = n(),
    n_studies = factor(n_distinct(study), levels = c(1:4)),
    .by = c(trt_group, analysis_plan, prereg_chr)
  ) |> 
  mutate(
    n_prereg = case_match(
      prereg_chr,
      "Preregistered" ~ 0.5,
      "Not preregistered" ~ 0
    ),
    
    constant = "Intervention",
    
    outcome = analysis_plan,
    
    intervention = case_match(
      trt_group,
      "group-based CBT" ~ "GrpCBT",
      "stress management" ~ "StressMan",
      "vocational training" ~ "JobTrain",
      "group psychotherapy"  ~ "GrpPsychT",
      "art therapy" ~ "ArtT",
      "illness management" ~ "IllMan",
      "group psychoeducation" ~ "GrpPsychEdu",
      "seeking safety" ~ "SeekSafe",
      "social cognition and interaction training" ~ "SCIT",
      "illness and addiction management" ~ "Ill&AddMan",
      "social network training" ~ "NetworkTrain",
      "cognitive-behavioral social skills training" ~ "CBSST",
      "residential treatment" ~ "ResidentTreat",
      "positive psychology group intervention" ~ "PosiPsychGrp",
      "addiction management" ~ "AddMan",
      "reading group intervention" ~ "ReadGrp",
      "group psychoeducation & social skill training" ~ "GrpPsyEdu&SS"
    )
    
    
    #intervention_name = case_match(
    #  intervention,
    #  "Sanctions and economic incentives" ~ "Econ incentives",
    #  "Introduction programmes" ~ "Intro program",
    #  "Combination programmes" ~ "Combi program",
    #  "Language training" ~ "Lang train",
    #  "Labour market training" ~ "Lab market train",
    #  "Job search assistance" ~" Job search ass",
    #  "Subsidized public sector employment" ~ "Sub public emp"
    #),
    #
    #outcome = if_else(outcome == "Duration of social assistance spells", "Dur. social ass spells", outcome)
    
  ) 

egm_dat_mental |> 
  ggplot(aes(x = n_prereg, y =  constant, size = n, color = n_studies)) + 
  geom_point() +
  scale_size(range = c(6, 14)) +
  geom_text(aes(label = n), color = "black", size = 3.5) +
  scale_x_continuous(breaks = seq(0,0.5,0.5), limits = c(0, 1), labels = c("Conventional", "Preregistered")) +
  scale_y_discrete("Interventions") +
  facet_grid(intervention~outcome, scales = "free_y", space = "free_y") +
  theme_bw() + 
  theme(
    legend.position = "bottom",
    axis.text.x = element_text(angle = 45, hjust = 1),
    #panel.grid.major = element_blank(),
    #panel.grid.minor = element_blank(),
    #panel.background = element_blank(),
    axis.text.y = element_blank(),   
    axis.ticks.y = element_blank(),  
    axis.title.y = element_blank(),
    plot.caption = element_text(hjust = 0, size = 12)
  ) +
  coord_cartesian(xlim = c(-0.25, 0.75), ylim = NULL) +
  guides(size = "none") + 
  xlab("Type of registration") +
  labs(
    color = "Number of studies",
    caption = paste0(
      "Note: GrpCBT = Group-based CBT, StressMan = Stress mangement, JobTrain = Vocational training,",
      "GrpPsychT = Group psychotherapy, ArtT = Art Therapy,\nIllMan = Ill management,",
      "GrpPsychEdu = Group Psychoeducation, SeekSafe = Seeking safety,",
      "SCIT = cognitive-behavioral social skills training,\nResidentTreat = Residential treatment,",
      "PosiPsychGrp = Positive Psychology group therapy, AddMan = Addiction mangement, ",
      "ReadGrp = Reading group,\nGrpPsyEdu&SS = Group psychoeducation & social skill training."
    )
  )

Figure 50: EGM for mental health outcomes

Type of test

test_type_dat <- 
  reintegration_dat |> 
  filter(test_type != "Raw events")

test_type_dat_cross <- cat_dat_cross(
  data = test_type_dat,
  variable = test_type,
  study_id = study
)

test_type_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Type of test",
      colnames(test_type_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 19: Dependency table for type of test (reintegration)

Type of test	Clinician-rated measure	Self-reported
Clinician-rated measure	12 (36)	5 (8)
Self-reported	5 (8)	38 (165)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

test_type_dat_mental <- 
  mental_health_dat |> 
  filter(test_type != "Raw events")


test_type_dat_cross_mental <- cat_dat_cross(
  data = test_type_dat_mental,
  variable = test_type,
  study_id = study
)

test_type_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Type of test",
      colnames(test_type_dat_cross_mental)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 20: Dependency table for type of test (mental health)

Type of test	Clinician-rated measure	Self-reported
Clinician-rated measure	14 (41)	5 (11)
Self-reported	5 (18)	32 (100)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cat_ridge(data = test_type_dat, es = gt_pop, variable = test_type, v = vgt_pop)

Figure 51: Distribution of effect size estimates, by type of test (reintegrational outcome).

cat_ridge(data = test_type_dat_mental, es = gt_pop, variable = test_type, v = vgt_pop)

Figure 52: Distribution of effect size estimates, by type of test (mental health).

test_type_dat_all <- 
  gb_dat |> 
  filter(test_type != "Raw events") |> 
  mutate(
    p_val = 2 * ( 1 - pnorm( abs(gt_pop) / sqrt(vgt_pop) ) ),
  )


ggplot(test_type_dat_all, aes(y = gt_pop, x = p_val, color = test_type)) +
geom_point() + 
geom_hline(yintercept = 0) + 
geom_vline(xintercept = .05, color = "gray") +
facet_grid(outcome_construct~test_type) +
theme_bw() +
theme(legend.position="none") +
labs(x = "p values", y = "Effect sizes (Hedges' g)")

Figure 53: Distributions of effect size by p values and outcome measure type.

ITT vs. TOT

strategy_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = analysis_strategy,
  study_id = study
)

strategy_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Strategy",
      colnames(strategy_subgroup_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses. ",
                     "TOT = treatment on the treated, ITT = Intention to treat."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 21: Dependency table for estimation strategy (reintegration)

Strategy	ITT	TOT
ITT	22 (108)	-
TOT	-	23 (94)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses. TOT = treatment on the treated, ITT = Intention to treat.

strategy_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = analysis_strategy,
  study_id = study
)

strategy_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Strategy",
      colnames(strategy_subgroup_dat_cross_mental)[-1]
    ) 
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses. ",
                     "TOT = treatment on the treated, ITT = Intention to treat."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  )  |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 22: Dependency table for estimation strategy (mental health)

Strategy	ITT	TOT
ITT	18 (69)	-
TOT	-	23 (72)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses. TOT = treatment on the treated, ITT = Intention to treat.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = analysis_strategy, v = vgt_pop)

Figure 54: Distribution of effect size estimates, by estiamtion strategy (reintegrational outcome).

cat_ridge(data = mental_health_dat, es = gt_pop, variable = analysis_strategy, v = vgt_pop)

Figure 55: Distribution of effect size estimates, by estiamtion strategy (mental health).

RCT vs. QES

design_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = QES_design,
  study_id = study
)

design_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Design",
      colnames(design_subgroup_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 23: Dependency table for type of research design (reintegration)

Design	QES	RCT
QES	8 (18)	-
RCT	-	37 (184)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

design_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = QES_design,
  study_id = study
)

design_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Design",
      colnames(design_subgroup_dat_cross_mental)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  )  |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 24: Dependency table for type of research design (mental health)

Design	QES	RCT
QES	8 (17)	-
RCT	-	33 (124)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = QES_design, v = vgt_pop)

Figure 56: Distribution of effect size estimates, by research design (reintegrational outcome).

cat_ridge(data = mental_health_dat, es = gt_pop, variable = QES_design, v = vgt_pop)

Figure 57: Distribution of effect size estimates, by research design (mental health).

Type of control

control_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = control,
  study_id = study
)

control_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Control",
      colnames(control_subgroup_dat_cross)[-1]
    ) 
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    paste0(
      "Values outside the parentheses are number of studies,", "
       with the number of samples and effect size estimates shown in the parentheses."
    ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 25: Dependency table for type of control group (reintegration)

Control	Individual treatment	TAU	TAU with Waiting-list	Waiting-list only
Individual treatment	3 (4)	-	-	-
TAU	-	30 (154)	-	-
TAU with Waiting-list	-	-	8 (31)	-
Waiting-list only	-	-	-	4 (13)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

control_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = control,
  study_id = study
)

control_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Control",
      colnames(control_subgroup_dat_cross_mental)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = "See note in the reintegration table to the left.", 
    general_title = "Note: ",
    footnote_as_chunk = T
  )  |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 26: Dependency table for type of control group (mental health)

Control	Individual treatment	TAU	TAU with Waiting-list	Waiting-list only
Individual treatment	2 (3)	-	-	-
TAU	-	28 (108)	-	-
TAU with Waiting-list	-	-	8 (23)	-
Waiting-list only	-	-	-	3 (7)
Note: See note in the reintegration table to the left.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = control, v = vgt_pop)

Figure 58: Distribution of effect size estimates, by type of control group (reintegrational outcome).

cat_ridge(data = mental_health_dat, es = gt_pop, variable = control, v = vgt_pop)

Figure 59: Distribution of effect size estimates, by type of control group (mental health).

Risk of bias (RoB)

rob_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = overall_rob,
  study_id = study
)

rob_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Risk of bias judgment",
      colnames(rob_subgroup_dat_cross)[-1]
    ) 
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 27: Dependency table across overall risk of bias assessments (reintegration)

Risk of bias judgment	Low	Some concerns/Moderate	Serious/High
Low	10 (69)	-	-
Some concerns/Moderate	-	21 (94)	1 (1)
Serious/High	-	1 (1)	15 (39)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

rob_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = overall_rob,
  study_id = study
)

rob_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Risk of bias judgment",
      colnames(rob_subgroup_dat_cross_mental)[-1]
    ) 
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  )  |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 28: Dependency table across overall risk of bias assessments (mental health)

Risk of bias judgment	Low	Some concerns/Moderate	Serious/High
Low	10 (36)	-	-
Some concerns/Moderate	-	20 (79)	1 (2)
Serious/High	-	1 (2)	12 (26)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = overall_rob, v = vgt_pop)

Figure 60: Distribution of effect size estimates, by overall risk of bias assessment (reintegrational outcome).

cat_ridge(data = gb_dat, es = gt_pop, variable = overall_rob, v = vgt_pop) 

Figure 61: Distribution of effect size estimates, by overall risk of bias assessment (mental health).

Preregistered vs. not preregistered studies

prereg_subgroup_dat_cross <- cat_dat_cross(
  data = reintegration_dat,
  variable = prereg_chr,
  study_id = study
)

prereg_subgroup_dat_cross |>
  knitr::kable(
    "html",
    col.names = c(
      "Registration",
      colnames(prereg_subgroup_dat_cross)[-1]
    )
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  ) |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 29: Dependency table by type of registration (reintegration)

Registration	Not preregistered	Preregistered
Not preregistered	23 (62)	-
Preregistered	-	22 (140)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

prereg_subgroup_dat_cross_mental <- cat_dat_cross(
  data = mental_health_dat,
  variable = prereg_chr,
  study_id = study
)

prereg_subgroup_dat_cross_mental |>
  knitr::kable(
    "html",
    col.names = c(
      "Risk of bias judgment",
      colnames(prereg_subgroup_dat_cross_mental)[-1]
    ) 
  ) |>
  kableExtra::column_spec(1, bold = TRUE) |>
  kableExtra::footnote(
    general = paste0("Values outside the parentheses are number of studies,", "
                     with the number of samples and effect size estimates shown in the parentheses."
                     ), 
    general_title = "Note: ",
    footnote_as_chunk = T
  )  |>
  kableExtra::kable_styling(bootstrap_options = c("striped", "condensed"), full_width = T)

Table 30: Dependency table by type of registration (mental health)

Risk of bias judgment	Not preregistered	Preregistered
Not preregistered	21 (59)	-
Preregistered	-	20 (82)
Note: Values outside the parentheses are number of studies, with the number of samples and effect size estimates shown in the parentheses.

cat_ridge(data = reintegration_dat, es = gt_pop, variable = prereg_chr, v = vgt_pop)

Figure 62: Distribution of effect size estimates, by type of registration (reintegrational outcome).

cat_ridge(data = gb_dat, es = gt_pop, variable = prereg_chr, v = vgt_pop) 

Figure 63: Distribution of effect size estimates, by type of registration (mental health).

cat_ridge(data = gb_dat, es = gt_pop, variable = overall_rob, v = vgt_pop) + 
  facet_grid(outcome_construct~prereg_chr) + 
  theme(
    strip.background = element_rect(color = "black", fill = "gray92")
  )

Figure 64: Distribution of effect size estimates, by risk of bias assessemnt across outcomes and registration (mental health).

Continuous moderators

var_labels <- c(
  "Mean age" = "age",
  "Percent Male" = "male_pct",
  "Total Number of Sessions" = "sessions",
  "Sessions per Week" = "intensity",
  "Length of Intervention (in Weeks)" = "duration",
  "Weeks After End of Intervention" = "timing",
  "Weeks from Baseline" = "weeks_from_baseline"
)

continuous_descriptives <- 
 reintegration_dat |> 
  summarise(
    age = mean(age_mean),
    male_pct = mean(male_pct),
    sessions = mean(total_number_of_sessions),
    intensity = mean(sessions_per_week),
    duration = mean(duration_in_weeks),
    timing = mean(time_after_end_intervention_weeks),
    weeks_from_baseline = mean(time_from_baseline_weeks),
    .by = study
  ) 

continuous_descriptives_tab <- 
  continuous_descriptives |> 
  #pivot_longer(
  #  cols = age:male_pct,
  #  names_to = "var",
  #  values_to = "val"
  #) |> 
  #arrange(var)
  gather(var, val, age, male_pct, sessions, intensity, duration, timing, weeks_from_baseline) |>
  summarise(
    `% Missing` = 100 * mean(is.na(val)),
    Mean = mean(val, na.rm = TRUE),
    SD = sd(val, na.rm = TRUE),
    Min = min(val, na.rm = TRUE),
    LQ = quantile(val, na.rm = TRUE)[2],
    Median = median(val, na.rm = TRUE),
    UQ = quantile(val, na.rm = TRUE)[4],
    Max = max(val, na.rm = TRUE),
    .by = var
  ) |> 
  mutate(
    var = factor(var, levels = var_labels, labels = names(var_labels))
  )

knitr::kable(
  continuous_descriptives_tab, 
  col.names = c("Variable", colnames(continuous_descriptives_tab)[-1]),
  digits = 1,
  booktabs = TRUE
) |>
  kable_styling(bootstrap_options = c("striped", "condensed"), full_width = FALSE) |>
  collapse_rows(1, valign = "top")

Table 31: Distribution of continuous moderators (reintegration)

Variable	% Missing	Mean	SD	Min	LQ	Median	UQ	Max
Mean age	0.0	40.7	9.1	24.9	35.6	40.8	43.8	67.4
Percent Male	2.2	45.8	23.0	0.0	31.5	45.7	67.4	79.5
Total Number of Sessions	2.2	21.2	22.0	3.0	8.0	12.0	24.0	104.0
Sessions per Week	2.2	1.3	1.5	0.1	1.0	1.0	1.0	10.5
Length of Intervention (in Weeks)	0.0	18.6	17.0	4.0	8.0	12.0	24.0	78.0
Weeks After End of Intervention	0.0	6.0	10.2	0.0	0.0	1.0	6.5	52.0
Weeks from Baseline	0.0	24.6	22.8	6.0	12.0	18.0	26.0	130.0

var_labels_mental <- c(
  "Mean age" = "age",
  "Percent Male" = "male_pct",
  "N Sessions" = "sessions",
  "Sessions Week" = "intensity",
  "Length" = "duration",
  "Weeks After Intervention" = "timing",
  "Weeks Baseline" = "weeks_from_baseline"
)

continuous_descriptives_mental <- 
 mental_health_dat |> 
  summarise(
    age = mean(age_mean),
    male_pct = mean(male_pct),
    sessions = mean(total_number_of_sessions),
    intensity = mean(sessions_per_week),
    duration = mean(duration_in_weeks),
    timing = mean(time_after_end_intervention_weeks),
    weeks_from_baseline = mean(time_from_baseline_weeks),
    .by = study
  ) 

continuous_descriptives_tab_mental <- 
  continuous_descriptives_mental |> 
  #pivot_longer(
  #  cols = age:male_pct,
  #  names_to = "var",
  #  values_to = "val"
  #) |> 
  #arrange(var)
  gather(var, val, age, male_pct, sessions, intensity, duration, timing, weeks_from_baseline) |>
  summarise(
    `% Missing` = 100 * mean(is.na(val)),
    Mean = mean(val, na.rm = TRUE),
    SD = sd(val, na.rm = TRUE),
    Min = min(val, na.rm = TRUE),
    LQ = quantile(val, na.rm = TRUE)[2],
    Median = median(val, na.rm = TRUE),
    UQ = quantile(val, na.rm = TRUE)[4],
    Max = max(val, na.rm = TRUE),
    .by = var
  ) |> 
  mutate(
    var = factor(var, levels = var_labels_mental, labels = names(var_labels_mental))
  )

knitr::kable(
  continuous_descriptives_tab_mental, 
  col.names = c("Variable", colnames(continuous_descriptives_tab)[-1]),
  digits = 1,
  booktabs = TRUE
) |>
  kable_styling(bootstrap_options = c("striped", "condensed"), full_width = FALSE) |>
  collapse_rows(1, valign = "top")

Table 32: Distribution of continuous moderators (mental health)

Variable	% Missing	Mean	SD	Min	LQ	Median	UQ	Max
Mean age	0.0	39.6	9.4	21.6	34.8	39.8	43.1	67.4
Percent Male	2.4	44.3	23.1	0.0	28.7	45.7	64.1	77.8
N Sessions	2.4	22.0	22.7	3.0	9.5	13.0	24.0	104.0
Sessions Week	2.4	1.3	1.6	0.1	1.0	1.0	1.0	10.5
Length	0.0	19.6	17.6	4.0	10.0	12.0	26.0	78.0
Weeks After Intervention	0.0	7.0	11.1	0.0	0.0	1.0	8.0	52.0
Weeks Baseline	0.0	26.6	24.3	6.0	12.0	18.0	29.0	130.0

Age distribution across studies

density_plot <- function(variable, x_title, data, color = "cornflowerblue") {
  require(dplyr)
  require(tidyr)
  require(ggplot2)
  require(rlang)
  require(MetBrewer)
  
  var_exp <- enquo(variable)
  var_str <- as_label(var_exp)
  
  data |>
    ggplot(aes(x = !!var_exp)) +
    geom_density(fill = color, alpha = 0.8) +
    geom_rug(alpha = 0.7, length = unit(0.04, "npc")) +
    scale_y_continuous(labels = scales::label_number(accuracy = 10^(-3))) +
    theme_minimal() +
    labs(x = x_title, y = "")
}

age_density <- density_plot(age, "Mean Age", continuous_descriptives)

age_density + expand_limits(x = 70)

Figure 65: Distribution of average age in a study (reintegrational).

age_density_mental <- density_plot(age, "Mean Age", continuous_descriptives_mental, color = "gray")

age_density_mental + expand_limits(x = 70)

Figure 66: Distribution of average age in a study (mental health).

Percent males in sample distribution across studies

male_density <- suppressWarnings(density_plot(male_pct, "Proportion Male", continuous_descriptives))
male_density 

Figure 67: Distribution proportion of males in each study (reintegration).

male_density_mental <- suppressWarnings(density_plot(male_pct, "Proportion Male", continuous_descriptives_mental, color = "gray"))
male_density_mental 

Figure 68: Distribution proportion of males in each study (mental health).

Total number of sessions

hist_plot <- function(variable, x_title, data, color = "cornflowerblue") {
  require(dplyr)
  require(tidyr)
  require(ggplot2)
  require(rlang)
  require(MetBrewer)
  
  var_exp <- enquo(variable)
  var_str <- as_label(var_exp)
  
  x_vals <- data[[var_str]]
  min_val <- floor(min(x_vals, na.rm = TRUE))
  max_val <- ceiling(max(x_vals, na.rm = TRUE))
  
  data |>
    ggplot(aes(x = !!var_exp)) +
    geom_histogram(binwidth = 1, boundary = 0, fill = color, alpha = 0.8) +
    scale_x_continuous(breaks = pretty(seq(min_val, max_val, by = 1), n = 5)) +
    scale_y_continuous(breaks = seq(0, 10, 2)) +
    theme_minimal() +
    labs(x = x_title, y = "")
}

sessions_hist <- hist_plot(sessions, "Total Number of Sessions in Intervention", continuous_descriptives)
sessions_hist + labs(title = "Reintegration") + theme(plot.title = element_text(hjust = 0.5))

Figure 69: Distribution of study sessions (reintegration).

sessions_hist_mental <- hist_plot(sessions, "Total Number of Sessions in Intervention", continuous_descriptives_mental, color = "gray")
sessions_hist_mental + labs(title = "Mental Health") + theme(plot.title = element_text(hjust = 0.5))

Figure 70: Distribution of study sessions (mental health).

Duration and intensity (number of sessions per week)

reintegration_dat |>
  select(study, duration_in_weeks, sessions_per_week, N_total) |> 
  filter(!is.na(sessions_per_week)) |>  
  arrange(desc(duration_in_weeks)) |> 
  mutate(
    study = factor(study, levels = unique(study)),
    sessions = case_when(
      sessions_per_week < 1 ~ "1<",
      sessions_per_week == 1 ~ "1",
      sessions_per_week == 1.5 ~ "1.5",
      sessions_per_week == 2 ~ "2",
      sessions_per_week == 2.5 ~ "2.5",
      sessions_per_week > 2 ~ ">10"
    ),
    sessions = factor(sessions, levels = unique(sessions))
  ) |> 
  ggplot(aes(y = study, x = duration_in_weeks, color = study)) +
  geom_segment(aes(x = 0, xend = duration_in_weeks, y = study, yend = study)) +
  geom_vline(xintercept = c(13, 26, 52, 78), linetype = "dashed", alpha = 0.5) +
  geom_point(
    aes(size = N_total)
  ) +
  scale_x_continuous(breaks = seq(0, 80, 10)) +
  facet_grid(vars(sessions_per_week), scales = "free_y", space = "free_y", switch = "y") + 
  #scale_size(breaks = c(0.5, 1, 1.5, 2, 10.5), limits = c(0, 10.5)) +
  labs(
    x = "Lenght of intervention in weeks", 
    y = "", 
    caption = paste0(
      "The light gray facet grids indicate the average number of sessions per week.\n", 
      "The dashed lines indicate 3 months, 6 months, 1 year, and 1.5 years, respectively.\n",
      "The point sizes are weighted by the study sample sizes."
      )
  ) +
  theme_minimal() + 
  scale_colour_discrete(guide = "none") +
  theme(
    legend.position = "none", 
    strip.text = element_text(color = "black"),
    strip.background.y = element_rect(fill = "gray93", color = "white"),
    strip.text.y.left = element_text(angle = 0),
    plot.caption = element_text(hjust = 0)
  ) 

Figure 71: Length of intervention in weeks (reintegration)

mental_health_dat |>
  select(study, duration_in_weeks, sessions_per_week, N_total) |> 
  filter(!is.na(sessions_per_week)) |>  
  arrange(desc(duration_in_weeks)) |> 
  mutate(
    study = factor(study, levels = unique(study)),
    sessions = case_when(
      sessions_per_week < 1 ~ "Less than 1",
      sessions_per_week == 1 ~ "1 per week",
      sessions_per_week == 2 ~ "2 per week",
      sessions_per_week > 2 ~ ">10"
    ),
    sessions = factor(sessions, levels = unique(sessions))
  ) |> 
  ggplot(aes(y = study, x = duration_in_weeks, color = study)) +
  geom_segment(aes(x = 0, xend = duration_in_weeks, y = study, yend = study)) +
  geom_vline(xintercept = c(13, 26, 52, 78), linetype = "dashed", alpha = 0.5) +
  geom_point(
    aes(size = N_total)
  ) +
  scale_x_continuous(breaks = seq(0, 80, 10)) +
  facet_grid(vars(sessions_per_week), scales = "free_y", space = "free_y", switch = "y") + 
  #scale_size(breaks = c(0.5, 1, 1.5, 2, 10.5), limits = c(0, 10.5)) +
  labs(
    x = "Lenght of intervention in weeks", 
    y = "", 
    caption = 
      paste0(
      "The light gray facet grids indicate the average number of sessions per week.\n", 
      "The dashed lines indicate 3 months, 6 months, 1 year, and 1.5 years, respectively.\n",
      "The point sizes are weighted by the study sample sizes."
      )
  ) +
  theme_minimal() + 
  scale_colour_discrete(guide = "none") +
  theme(
    legend.position = "none", 
    strip.text = element_text(color = "black"),
    strip.background.y = element_rect(fill = "gray93", color = "white"),
    strip.text.y.left = element_text(angle = 0),
    plot.caption = element_text(hjust = 0)
  )  

Figure 72: Length of intervention in weeks (mental health)

Weeks after end of intervention

reintegration_dat |>
  select(study, time_after_end_intervention_weeks, N_total) |> 
  arrange(desc(study)) |> 
  mutate(study = factor(study, levels = unique(study))) |> 
  ggplot(aes(y = study, x = time_after_end_intervention_weeks, color = study)) +
  geom_point(aes(size = N_total),
             alpha = 0.5) +
  geom_line() +
  labs(x = "Weeks after end of intervention all studies", y = "")+
  theme_minimal() + 
  theme(legend.position = "none")

Figure 73: Weeks after end of interventions for all studies (reintegration)

mental_health_dat |>
  select(study, time_after_end_intervention_weeks, N_total) |> 
  arrange(desc(study)) |> 
  mutate(study = factor(study, levels = unique(study))) |> 
  ggplot(aes(y = study, x = time_after_end_intervention_weeks, color = study)) +
  geom_point(aes(size = N_total),
             alpha = 0.5) +
  geom_line() +
  labs(x = "Weeks after end of intervention all studies", y = "")+
  theme_minimal() + 
  theme(legend.position = "none")

Figure 74: Weeks after end of interventions for all studies (mental health)

Correlation between effect size estimates and measurement timing

reintegration_dat |> 
  select(study, gt_pop, vgt_pop, time_after_end_intervention_weeks) |> 
  ggplot() +
  aes(x = time_after_end_intervention_weeks, y = gt_pop, color = study) +
  geom_vline(xintercept = 0) + 
  geom_hline(yintercept = 0) + 
  geom_point(aes(size = 1/vgt_pop), alpha = 0.30) + 
  geom_smooth(method = "lm", formula = y ~ x, color = "yellow") + 
  scale_x_continuous(breaks = seq(0, 55, 5)) +
  theme_minimal() + 
  theme(legend.position = "none") + 
  labs(x = "Follow-up duration (months)", y = "Hedges' g")

Figure 75: Length of intervention in weeks (reintegration)

mental_health_dat |> 
  select(study, gt_pop, vgt_pop, time_after_end_intervention_weeks) |> 
  ggplot() +
  aes(x = time_after_end_intervention_weeks, y = gt_pop, color = study) +
  geom_vline(xintercept = 0) + 
  geom_hline(yintercept = 0) + 
 geom_point(aes(size = 1/vgt_pop), alpha = 0.30) + 
  geom_smooth(method = "lm", formula = y ~ x, color = "yellow") + 
  scale_x_continuous(breaks = seq(0, 55, 5)) +
  theme_minimal() + 
  theme(legend.position = "none") + 
  labs(x = "Follow-up duration (months)", y = "Hedges' g")

Figure 76: Length of intervention in weeks (mental health)

Multivariate structure

Covariance plots

Substantial categorical and continuous moderators

Reintegration (across subgroup analyzed outcomes)

multivariate_dat_reint <- 
  reintegration_dat |>
  filter(str_detect(analysis_plan, "Alco|Hope|Social|Well")) |> 
  select(
    `Outcome` = analysis_plan,
    `Sample diagnosis` = schizophrenia,
    `Intervention` = CBT_int,
    `Type of test` = test_type,
    `Mean age` = age_mean,
    `% male` = male_pct,
    `Duration` = duration_in_weeks,
    `Intensity` = sessions_per_week,
    `Follow-up` = time_after_end_intervention_weeks
  )

multivariate_pairs_reint <- ggpairs(multivariate_dat_reint) + theme(
    text = element_text(size = 18),
    axis.text.x = element_text(size = 10, angle = 45, vjust = 1, hjust=1),  
    axis.text.y = element_text(size = 10) 
  ) 

multivariate_pairs_reint

Figure 77: Multivariate structure between substaintial/theoretical categorical and continuous (reintegration)

Mental health

multivariate_dat_mental <- 
  mental_health_dat |> 
  select(
    `Outcome` = analysis_plan,
    `Sample diagnosis` = schizophrenia,
    `Intervention` = CBT_int,
    `Type of test` = test_type,
    `Mean age` = age_mean,
    `% male` = male_pct,
    `Duration` = duration_in_weeks,
    `Intensity` = sessions_per_week,
    `Follow-up` = time_after_end_intervention_weeks
  )

multivariate_pairs_mental <- ggpairs(multivariate_dat_mental) + theme(
    text = element_text(size = 18),
    axis.text.x = element_text(size = 10, angle = 45, vjust = 1, hjust=1),  
    axis.text.y = element_text(size = 10) 
  )

multivariate_pairs_mental

Figure 78: Multivariate structure between substaintial/theoretical categorical and continuous covariates (mental health)

Methodological categorical and continuous moderators

Reintegration (across subgroup analyzed outcomes)

multivariate_dat_reint_method <- 
  reintegration_dat |>
  filter(str_detect(analysis_plan, "Alco|Hope|Social|Well")) |> 
  select(
    `Strategy` = analysis_strategy,
    `Design` = QES_design,
    `Control grp` = control,
    `RoB` = overall_rob,
    `Preregistration` = prereg_chr,
    `Mean age` = age_mean,
    `% male` = male_pct,
    `Duration` = duration_in_weeks,
    `Intensity` = sessions_per_week,
    `Follow-up` = time_after_end_intervention_weeks
  )

multivariate_pairs_reint_method <- ggpairs(multivariate_dat_reint_method) + theme(
    text = element_text(size = 18),
    axis.text.x = element_text(size = 9, angle = 45, vjust = 1, hjust=1),  
    axis.text.y = element_text(size = 9) 
  ) 

multivariate_pairs_reint_method

Figure 79: Multivariate structure between methodological categorical and continuous covariates (reintegration)

Mental health

multivariate_dat_mental_method <- 
  mental_health_dat |> 
  select(
    `Strategy` = analysis_strategy,
    `Design` = QES_design,
    `Control grp` = control,
    `RoB` = overall_rob,
    `Preregistration` = prereg_chr,
    `Mean age` = age_mean,
    `% male` = male_pct,
    `Duration` = duration_in_weeks,
    `Intensity` = sessions_per_week,
    `Follow-up` = time_after_end_intervention_weeks
  )

multivariate_pairs_mental_method <- ggpairs(multivariate_dat_mental_method) + theme(
    text = element_text(size = 18),
    axis.text.x = element_text(size = 9, angle = 45, vjust = 1, hjust=1),  
    axis.text.y = element_text(size = 9) 
  )

multivariate_pairs_mental_method 

Figure 80: Multivariate structure between methodological categorical and continuous (mental health)

Substantial and methodological categorical moderators

Reintegration (across subgroup analyzed outcomes)

multivariate_dat_reint_method_theo <- 
  reintegration_dat |>
  filter(str_detect(analysis_plan, "Alco|Hope|Social|Well")) |> 
  select(
    `Outcome` = analysis_plan,
    `Sample diagnosis` = schizophrenia,
    `Intervention` = CBT_int,
    `Type of test` = test_type,
    `Strategy` = analysis_strategy,
    `Design` = QES_design,
    `Control grp` = control,
    `RoB` = overall_rob,
    `Preregistration` = prereg_chr
  )

multivariate_pairs_reint_method_theo <- ggpairs(multivariate_dat_reint_method_theo) + theme(
    text = element_text(size = 18),
    axis.text.x = element_text(size = 9, angle = 45, vjust = 1, hjust=1),  
    axis.text.y = element_text(size = 9) 
  ) 

multivariate_pairs_reint_method_theo

Figure 81: Multivariate structure between substantial and methodological categorical covariates (reintegration)

Mental health

multivariate_dat_mental_method_theo <- 
  mental_health_dat |> 
  select(
    `Outcome` = analysis_plan,
    `Sample diagnosis` = schizophrenia,
    `Intervention` = CBT_int,
    `Type of test` = test_type,
    `Strategy` = analysis_strategy,
    `Design` = QES_design,
    `Control grp` = control,
    `RoB` = overall_rob,
    `Preregistration` = prereg_chr,
    `Mean age` = age_mean,
    `% male` = male_pct,
    `Duration` = duration_in_weeks,
    `Intensity` = sessions_per_week,
    `Follow-up` = time_after_end_intervention_weeks
  )

multivariate_pairs_mental_method_theo <- ggpairs(multivariate_dat_mental_method_theo) + theme(
    text = element_text(size = 18),
    axis.text.x = element_text(size = 9, angle = 45, vjust = 1, hjust=1),  
    axis.text.y = element_text(size = 9) 
  )

multivariate_pairs_mental_method_theo

Figure 82: Multivariate structure between substantial and methodological categorical covariates (mental health)

Covariance matrices

Reintegrational outcomes

Across all outcomes

cor_mat_dat_cat <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Alco|Hope|Social|Well") & test_type != "Raw events") |> 
  select(
    plan = analysis_plan,
    samp = schizophrenia,
    treat = CBT_int,
    test = test_type,
    str = analysis_strategy,
    des = QES_design,
    ctr = control,
    rob = overall_rob,
    pre = prereg_chr
  )

cat_dummy_dat <- 
  fastDummies::dummy_cols(cor_mat_dat_cat) %>% 
  select(where(is.numeric))


cor_mat_dat_con <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Alco|Hope|Social|Well") & test_type != "Raw events") |> 
  select(
    age = age_mean,
    male = male_pct,
    dur = duration_in_weeks,
    int = sessions_per_week,
    FU = time_after_end_intervention_weeks,
    sess = total_number_of_sessions
  )

cor_mat_dat <- 
  bind_cols(cat_dummy_dat, cor_mat_dat_con) |> 
  select(
    alcho = `plan_Alcohol and drug abuse/misuse`,
    hope = `plan_Hope, empowerment & self-efficacy`,
    sfunc = `plan_Social functioning (degree of impairment)`,
    wellb = `plan_Wellbeing and quality of life`,
    s_schizo = samp_Schizophrenia,
    s_oth = samp_Other,
    cbt_trt = treat_CBT, 
    oth_trt = treat_Other,
    test_clin = `test_Clinician-rated measure`,
    test_self = `test_Self-reported`,
    itt = str_ITT,
    tot = str_TOT,
    qes = des_QES,
    rct = des_RCT,
    tau = ctr_TAU,
    tau_wait = `ctr_TAU with Waiting-list`,
    wait = `ctr_Waiting-list only`,
    ind_trt = `ctr_Individual treatment` ,
    rob_low = rob_Low,
    rob_mod = `rob_Some concerns/Moderate`,
    rob_high = `rob_Serious/High`,
    prereg = pre_Preregistered, 
    conventional = `pre_Not preregistered`,
    everything()
  ) |> 
  na.omit()

cor_mat <- 
  cor(
  model.matrix(
    reformulate(names(cor_mat_dat[,1:ncol(cor_mat_dat)])), 
    data = cor_mat_dat
    )
  ) |>  
  as.data.frame() |> 
  select(-c(`(Intercept)`)) |>  
  na.omit() |> 
  mutate(
    
    across(.cols = everything(), ~ round(.x, 2))
  )

cor_mat_formatted <- 
  cor_mat |> 
  mutate(
    across(.cols = everything(), ~ cell_spec(.x, bold = ifelse(abs(.x) > 0.5, T, F)))
  )

kbl(
  cor_mat_formatted, 
  escape = F,
  col.names = c("Category", colnames(cor_mat))
  ) |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "100%", fixed_thead = TRUE)

Table 33: Correlation matrix across all reintegration outcomes.

Category	alcho	hope	sfunc	wellb	s_schizo	s_oth	cbt_trt	oth_trt	test_clin	test_self	itt	tot	qes	rct	tau	tau_wait	wait	ind_trt	rob_low	rob_mod	rob_high	prereg	conventional	age	male	dur	int	FU	sess
alcho	1	-0.22	-0.27	-0.36	-0.19	0.19	-0.23	0.23	0.49	-0.49	0.33	-0.33	0.03	-0.03	0.04	0.03	-0.1	-0.06	-0.36	0.39	-0.05	0.02	-0.02	-0.26	-0.37	0.01	0.06	-0.05	0.14
hope	-0.22	1	-0.28	-0.38	0.02	-0.02	-0.31	0.31	-0.23	0.23	-0.16	0.16	-0.05	0.05	-0.16	-0.02	0.32	0.05	-0.22	0.17	0.06	-0.17	0.17	0.12	0.17	-0.14	-0.04	-0.13	-0.11
sfunc	-0.27	-0.28	1	-0.46	0.19	-0.19	0.23	-0.23	-0.04	0.04	-0.01	0.01	-0.08	0.08	0.06	0	-0.13	0.02	0.3	-0.26	-0.04	0.06	-0.06	-0.17	0.02	0.3	-0.06	0.19	0.17
wellb	-0.36	-0.38	-0.46	1	-0.04	0.04	0.22	-0.22	-0.16	0.16	-0.13	0.13	0.08	-0.08	0.04	-0.01	-0.06	-0.01	0.19	-0.2	0.03	0.07	-0.07	0.27	0.13	-0.16	0.04	-0.02	-0.17
s_schizo	-0.19	0.02	0.19	-0.04	1	-1	-0.26	0.26	-0.07	0.07	0.2	-0.2	-0.02	0.02	0.12	-0.16	-0.09	0.2	0.38	-0.36	-0.02	0.08	-0.08	-0.21	0.5	0.49	-0.01	0.06	0.38
s_oth	0.19	-0.02	-0.19	0.04	-1	1	0.26	-0.26	0.07	-0.07	-0.2	0.2	0.02	-0.02	-0.12	0.16	0.09	-0.2	-0.38	0.36	0.02	-0.08	0.08	0.21	-0.5	-0.49	0.01	-0.06	-0.38
cbt_trt	-0.23	-0.31	0.23	0.22	-0.26	0.26	1	-1	-0.11	0.11	-0.25	0.25	-0.03	0.03	0.13	-0.15	-0.02	0.01	0.31	-0.2	-0.15	0.1	-0.1	0.04	-0.32	-0.26	-0.04	0.2	-0.27
oth_trt	0.23	0.31	-0.23	-0.22	0.26	-0.26	-1	1	0.11	-0.11	0.25	-0.25	0.03	-0.03	-0.13	0.15	0.02	-0.01	-0.31	0.2	0.15	-0.1	0.1	-0.04	0.32	0.26	0.04	-0.2	0.27
test_clin	0.49	-0.23	-0.04	-0.16	-0.07	0.07	-0.11	0.11	1	-1	0.23	-0.23	0.09	-0.09	0.17	-0.11	-0.11	-0.06	-0.28	0.26	0.02	-0.03	0.03	-0.19	-0.37	-0.03	0.06	-0.02	0.03
test_self	-0.49	0.23	0.04	0.16	0.07	-0.07	0.11	-0.11	-1	1	-0.23	0.23	-0.09	0.09	-0.17	0.11	0.11	0.06	0.28	-0.26	-0.02	0.03	-0.03	0.19	0.37	0.03	-0.06	0.02	-0.03
itt	0.33	-0.16	-0.01	-0.13	0.2	-0.2	-0.25	0.25	0.23	-0.23	1	-1	-0.1	0.1	0.26	-0.1	-0.24	-0.15	-0.19	0.25	-0.1	0.17	-0.17	-0.15	-0.02	0.27	-0.02	-0.07	0.21
tot	-0.33	0.16	0.01	0.13	-0.2	0.2	0.25	-0.25	-0.23	0.23	-1	1	0.1	-0.1	-0.26	0.1	0.24	0.15	0.19	-0.25	0.1	-0.17	0.17	0.15	0.02	-0.27	0.02	0.07	-0.21
qes	0.03	-0.05	-0.08	0.08	-0.02	0.02	-0.03	0.03	0.09	-0.09	-0.1	0.1	1	-1	-0.01	-0.1	0.07	0.17	-0.18	-0.18	0.51	-0.41	0.41	-0.12	-0.06	0.22	0.41	-0.07	0.47
rct	-0.03	0.05	0.08	-0.08	0.02	-0.02	0.03	-0.03	-0.09	0.09	0.1	-0.1	-1	1	0.01	0.1	-0.07	-0.17	0.18	0.18	-0.51	0.41	-0.41	0.12	0.06	-0.22	-0.41	0.07	-0.47
tau	0.04	-0.16	0.06	0.04	0.12	-0.12	0.13	-0.13	0.17	-0.17	0.26	-0.26	-0.01	0.01	1	-0.8	-0.44	-0.26	0.1	0.11	-0.29	0.04	-0.04	0.02	-0.07	0.06	-0.3	0.28	-0.11
tau_wait	0.03	-0.02	0	-0.01	-0.16	0.16	-0.15	0.15	-0.11	0.11	-0.1	0.1	-0.1	0.1	-0.8	1	-0.09	-0.05	0	-0.26	0.37	0.2	-0.2	-0.11	0.03	-0.07	0.17	-0.24	0.07
wait	-0.1	0.32	-0.13	-0.06	-0.09	0.09	-0.02	0.02	-0.11	0.11	-0.24	0.24	0.07	-0.07	-0.44	-0.09	1	-0.03	-0.17	0.23	-0.09	-0.32	0.32	0.24	0	-0.08	0.3	-0.09	0.02
ind_trt	-0.06	0.05	0.02	-0.01	0.2	-0.2	0.01	-0.01	-0.06	0.06	-0.15	0.15	0.17	-0.17	-0.26	-0.05	-0.03	1	-0.01	-0.04	0.07	-0.13	0.13	-0.14	0.14	0.14	0	-0.08	0.11
rob_low	-0.36	-0.22	0.3	0.19	0.38	-0.38	0.31	-0.31	-0.28	0.28	-0.19	0.19	-0.18	0.18	0.1	0	-0.17	-0.01	1	-0.75	-0.32	0.45	-0.45	-0.01	0.12	0.36	-0.31	0.28	0.01
rob_mod	0.39	0.17	-0.26	-0.2	-0.36	0.36	-0.2	0.2	0.26	-0.26	0.25	-0.25	-0.18	0.18	0.11	-0.26	0.23	-0.04	-0.75	1	-0.39	-0.3	0.3	0.19	-0.04	-0.38	0.15	-0.19	-0.19
rob_high	-0.05	0.06	-0.04	0.03	-0.02	0.02	-0.15	0.15	0.02	-0.02	-0.1	0.1	0.51	-0.51	-0.29	0.37	-0.09	0.07	-0.32	-0.39	1	-0.19	0.19	-0.25	-0.11	0.05	0.21	-0.13	0.26
prereg	0.02	-0.17	0.06	0.07	0.08	-0.08	0.1	-0.1	-0.03	0.03	0.17	-0.17	-0.41	0.41	0.04	0.2	-0.32	-0.13	0.45	-0.3	-0.19	1	-1	0.34	-0.21	0.03	-0.27	0.13	-0.19
conventional	-0.02	0.17	-0.06	-0.07	-0.08	0.08	-0.1	0.1	0.03	-0.03	-0.17	0.17	0.41	-0.41	-0.04	-0.2	0.32	0.13	-0.45	0.3	0.19	-1	1	-0.34	0.21	-0.03	0.27	-0.13	0.19
age	-0.26	0.12	-0.17	0.27	-0.21	0.21	0.04	-0.04	-0.19	0.19	-0.15	0.15	-0.12	0.12	0.02	-0.11	0.24	-0.14	-0.01	0.19	-0.25	0.34	-0.34	1	-0.01	-0.29	0	-0.01	-0.24
male	-0.37	0.17	0.02	0.13	0.5	-0.5	-0.32	0.32	-0.37	0.37	-0.02	0.02	-0.06	0.06	-0.07	0.03	0	0.14	0.12	-0.04	-0.11	-0.21	0.21	-0.01	1	0.24	-0.01	-0.09	0.1
dur	0.01	-0.14	0.3	-0.16	0.49	-0.49	-0.26	0.26	-0.03	0.03	0.27	-0.27	0.22	-0.22	0.06	-0.07	-0.08	0.14	0.36	-0.38	0.05	0.03	-0.03	-0.29	0.24	1	-0.05	0.21	0.77
int	0.06	-0.04	-0.06	0.04	-0.01	0.01	-0.04	0.04	0.06	-0.06	-0.02	0.02	0.41	-0.41	-0.3	0.17	0.3	0	-0.31	0.15	0.21	-0.27	0.27	0	-0.01	-0.05	1	-0.19	0.41
FU	-0.05	-0.13	0.19	-0.02	0.06	-0.06	0.2	-0.2	-0.02	0.02	-0.07	0.07	-0.07	0.07	0.28	-0.24	-0.09	-0.08	0.28	-0.19	-0.13	0.13	-0.13	-0.01	-0.09	0.21	-0.19	1	0
sess	0.14	-0.11	0.17	-0.17	0.38	-0.38	-0.27	0.27	0.03	-0.03	0.21	-0.21	0.47	-0.47	-0.11	0.07	0.02	0.11	0.01	-0.19	0.26	-0.19	0.19	-0.24	0.1	0.77	0.41	0	1

Alcohol and drug abuse/misuse

Excluded factors due to no variation: type of sample (schizophania vs. rest), low risk of bias.

cor_mat_dat_cat_alcohol <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Alco") & test_type != "Raw events") |> 
  select(
    treat = CBT_int,
    test = test_type,
    str = analysis_strategy,
    des = QES_design,
    ctr = control,
    rob = overall_rob,
    pre = prereg_chr
  )

cat_dummy_dat_alcohol <- 
  fastDummies::dummy_cols(cor_mat_dat_cat_alcohol) %>% 
  select(where(is.numeric))


cor_mat_dat_con_alcohol <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Alco") & test_type != "Raw events") |> 
  select(
    age = age_mean,
    male = male_pct,
    dur = duration_in_weeks,
    int = sessions_per_week,
    FU = time_after_end_intervention_weeks,
    sess = total_number_of_sessions
  )

cor_mat_dat_alcohol <- 
  bind_cols(cat_dummy_dat_alcohol, cor_mat_dat_con_alcohol) |> 
  select(
    cbt_trt = treat_CBT, 
    oth_trt = treat_Other,
    test_clin = `test_Clinician-rated measure`,
    test_self = `test_Self-reported`,
    itt = str_ITT,
    tot = str_TOT,
    qes = des_QES,
    rct = des_RCT,
    tau = ctr_TAU,
    tau_wait = `ctr_TAU with Waiting-list`,
    rob_mod = `rob_Some concerns/Moderate`,
    rob_high = `rob_Serious/High`,
    prereg = pre_Preregistered, 
    conventional = `pre_Not preregistered`,
    age:sess
  ) |> 
  na.omit()

cor_mat_alcohol <- 
  cor(
    model.matrix(
      reformulate(names(cor_mat_dat_alcohol[,1:ncol(cor_mat_dat_alcohol)])), 
      data = cor_mat_dat_alcohol
    )
  ) |> 
  as.data.frame() |> 
  select(-c(`(Intercept)`)) |>   
  na.omit() |> 
  mutate(
    
    across(.cols = everything(), ~ round(.x, 2))
  )

cor_mat_formatted_alcohol <- 
  cor_mat_alcohol |> 
  mutate(
    across(.cols = everything(), ~ cell_spec(.x, bold = ifelse(abs(.x) > 0.5, T, F)))
  )


kbl(
  cor_mat_formatted_alcohol, 
  escape = F,
  col.names = c("Category", colnames(cor_mat_alcohol))
  ) |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "100%", fixed_thead = TRUE)

Table 34: Correlation matrix based on alcohol outcome only.

Category	cbt_trt	oth_trt	test_clin	test_self	itt	tot	qes	rct	tau	tau_wait	rob_mod	rob_high	prereg	conventional	age	male	dur	int	FU	sess
cbt_trt	1	-1	-0.33	0.33	0.09	-0.09	-0.07	0.07	0.12	-0.12	0.09	-0.09	-0.48	0.48	-0.5	0.16	-0.17	-0.12	0.27	-0.13
oth_trt	-1	1	0.33	-0.33	-0.09	0.09	0.07	-0.07	-0.12	0.12	-0.09	0.09	0.48	-0.48	0.5	-0.16	0.17	0.12	-0.27	0.13
test_clin	-0.33	0.33	1	-1	0.41	-0.41	-0.33	0.33	0.55	-0.55	0.41	-0.41	0.68	-0.68	0.63	-0.79	-0.28	-0.57	0.32	-0.4
test_self	0.33	-0.33	-1	1	-0.41	0.41	0.33	-0.33	-0.55	0.55	-0.41	0.41	-0.68	0.68	-0.63	0.79	0.28	0.57	-0.32	0.4
itt	0.09	-0.09	0.41	-0.41	1	-1	0.09	-0.09	-0.15	0.15	-0.11	0.11	0.6	-0.6	0.64	-0.59	0.41	0.15	0.1	0.25
tot	-0.09	0.09	-0.41	0.41	-1	1	-0.09	0.09	0.15	-0.15	0.11	-0.11	-0.6	0.6	-0.64	0.59	-0.41	-0.15	-0.1	-0.25
qes	-0.07	0.07	-0.33	0.33	0.09	-0.09	1	-1	0.12	-0.12	-0.8	0.8	-0.48	0.48	-0.24	-0.17	0.89	0.81	-0.23	0.94
rct	0.07	-0.07	0.33	-0.33	-0.09	0.09	-1	1	-0.12	0.12	0.8	-0.8	0.48	-0.48	0.24	0.17	-0.89	-0.81	0.23	-0.94
tau	0.12	-0.12	0.55	-0.55	-0.15	0.15	0.12	-0.12	1	-1	0.15	-0.15	-0.25	0.25	-0.18	-0.56	-0.22	-0.42	0.38	-0.18
tau_wait	-0.12	0.12	-0.55	0.55	0.15	-0.15	-0.12	0.12	-1	1	-0.15	0.15	0.25	-0.25	0.18	0.56	0.22	0.42	-0.38	0.18
rob_mod	0.09	-0.09	0.41	-0.41	-0.11	0.11	-0.8	0.8	0.15	-0.15	1	-1	0.34	-0.34	0.37	0.19	-0.76	-0.62	0.28	-0.76
rob_high	-0.09	0.09	-0.41	0.41	0.11	-0.11	0.8	-0.8	-0.15	0.15	-1	1	-0.34	0.34	-0.37	-0.19	0.76	0.62	-0.28	0.76
prereg	-0.48	0.48	0.68	-0.68	0.6	-0.6	-0.48	0.48	-0.25	0.25	0.34	-0.34	1	-1	0.89	-0.42	-0.14	-0.29	0.04	-0.3
conventional	0.48	-0.48	-0.68	0.68	-0.6	0.6	0.48	-0.48	0.25	-0.25	-0.34	0.34	-1	1	-0.89	0.42	0.14	0.29	-0.04	0.3
age	-0.5	0.5	0.63	-0.63	0.64	-0.64	-0.24	0.24	-0.18	0.18	0.37	-0.37	0.89	-0.89	1	-0.35	0.1	0	0	-0.03
male	0.16	-0.16	-0.79	0.79	-0.59	0.59	-0.17	0.17	-0.56	0.56	0.19	-0.19	-0.42	0.42	-0.35	1	-0.17	0.24	-0.25	-0.06
dur	-0.17	0.17	-0.28	0.28	0.41	-0.41	0.89	-0.89	-0.22	0.22	-0.76	0.76	-0.14	0.14	0.1	-0.17	1	0.91	-0.3	0.98
int	-0.12	0.12	-0.57	0.57	0.15	-0.15	0.81	-0.81	-0.42	0.42	-0.62	0.62	-0.29	0.29	0	0.24	0.91	1	-0.39	0.94
FU	0.27	-0.27	0.32	-0.32	0.1	-0.1	-0.23	0.23	0.38	-0.38	0.28	-0.28	0.04	-0.04	0	-0.25	-0.3	-0.39	1	-0.32
sess	-0.13	0.13	-0.4	0.4	0.25	-0.25	0.94	-0.94	-0.18	0.18	-0.76	0.76	-0.3	0.3	-0.03	-0.06	0.98	0.94	-0.32	1

Hope, empowerment & self-efficacy

Excluded factors due to no variation: type of intervention (CBT vs. rest) and type of test.

cor_mat_dat_cat_hope <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Hope") & test_type != "Raw events") |> 
  select(
    samp = schizophrenia,
    treat = CBT_int,
    test = test_type,
    str = analysis_strategy,
    des = QES_design,
    ctr = control,
    rob = overall_rob,
    pre = prereg_chr
  )

cat_dummy_dat_hope <- 
  fastDummies::dummy_cols(cor_mat_dat_cat_hope) %>% 
  select(where(is.numeric))


cor_mat_dat_con_hope <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Hope") & test_type != "Raw events") |> 
  select(
    age = age_mean,
    male = male_pct,
    dur = duration_in_weeks,
    int = sessions_per_week,
    FU = time_after_end_intervention_weeks,
    sess = total_number_of_sessions
  )

cor_mat_dat_hope <- 
  bind_cols(cat_dummy_dat_hope, cor_mat_dat_con_hope) |> 
  select(
    s_schizo = samp_Schizophrenia,
    s_oth = samp_Other,
    itt = str_ITT,
    tot = str_TOT,
    qes = des_QES,
    rct = des_RCT,
    tau = ctr_TAU,
    tau_wait = `ctr_TAU with Waiting-list`,
    wait = `ctr_Waiting-list only`,
    ind_trt = `ctr_Individual treatment` ,
    rob_low = rob_Low,
    rob_mod = `rob_Some concerns/Moderate`,
    rob_high = `rob_Serious/High`,
    prereg = pre_Preregistered, 
    conventional = `pre_Not preregistered`,
    age:sess
  ) |> 
  na.omit()

cor_mat_hope <- 
  cor(
    model.matrix(
      reformulate(names(cor_mat_dat_hope[,1:ncol(cor_mat_dat_hope)])), 
      data = cor_mat_dat_hope
    )
  ) |>  
  as.data.frame() |>  
  select(-c(`(Intercept)`)) |>  
  na.omit() |> 
  mutate(
    across(.cols = everything(), ~ round(.x, 2))
  )

cor_mat_formatted_hope <- 
  cor_mat_hope |> 
  mutate(
    across(.cols = everything(), ~ cell_spec(.x, bold = ifelse(abs(.x) > 0.5, T, F)))
  )

kbl(
  cor_mat_formatted_hope, 
  escape = F,
  col.names = c("Category", colnames(cor_mat_hope))
  ) |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "100%", fixed_thead = TRUE)

Table 35: Correlation matrix based on hope, empowerment & self-efficacy outcomes only.

Category	s_schizo	s_oth	itt	tot	qes	rct	tau	tau_wait	wait	ind_trt	rob_low	rob_mod	rob_high	prereg	conventional	age	male	dur	int	FU	sess
s_schizo	1	-1	0.2	-0.2	-0.08	0.08	0.13	-0.16	-0.21	0.42	0.53	-0.41	0.01	0.01	-0.01	-0.33	0.52	-0.1	0.03	-0.25	-0.05
s_oth	-1	1	-0.2	0.2	0.08	-0.08	-0.13	0.16	0.21	-0.42	-0.53	0.41	-0.01	-0.01	0.01	0.33	-0.52	0.1	-0.03	0.25	0.05
itt	0.2	-0.2	1	-1	0.23	-0.23	0.56	-0.29	-0.37	-0.14	0.56	-0.25	-0.21	-0.02	0.02	-0.3	0.07	0.43	-0.1	-0.06	0.15
tot	-0.2	0.2	-1	1	-0.23	0.23	-0.56	0.29	0.37	0.14	-0.56	0.25	0.21	0.02	-0.02	0.3	-0.07	-0.43	0.1	0.06	-0.15
qes	-0.08	0.08	0.23	-0.23	1	-1	0.13	-0.07	-0.09	-0.03	-0.08	-0.25	0.37	-0.22	0.22	-0.36	-0.51	0.64	0.35	-0.13	0.93
rct	0.08	-0.08	-0.23	0.23	-1	1	-0.13	0.07	0.09	0.03	0.08	0.25	-0.37	0.22	-0.22	0.36	0.51	-0.64	-0.35	0.13	-0.93
tau	0.13	-0.13	0.56	-0.56	0.13	-0.13	1	-0.52	-0.66	-0.25	0.31	0.03	-0.33	0.34	-0.34	-0.24	0.23	0.05	-0.56	0.09	-0.17
tau_wait	-0.16	0.16	-0.29	0.29	-0.07	0.07	-0.52	1	-0.18	-0.07	-0.16	-0.52	0.79	0.31	-0.31	0.02	-0.28	-0.06	0.75	-0.02	0.23
wait	-0.21	0.21	-0.37	0.37	-0.09	0.09	-0.66	-0.18	1	-0.09	-0.21	0.35	-0.23	-0.58	0.58	0.27	-0.21	0.01	0.04	-0.03	0.02
ind_trt	0.42	-0.42	-0.14	0.14	-0.03	0.03	-0.25	-0.07	-0.09	1	-0.08	0.13	-0.09	-0.22	0.22	0.02	0.37	-0.03	0.01	-0.13	-0.01
rob_low	0.53	-0.53	0.56	-0.56	-0.08	0.08	0.31	-0.16	-0.21	-0.08	1	-0.59	-0.21	0.36	-0.36	-0.22	0.34	0.36	-0.27	0.1	-0.14
rob_mod	-0.41	0.41	-0.25	0.25	-0.25	0.25	0.03	-0.52	0.35	0.13	-0.59	1	-0.66	-0.33	0.33	0.52	0.02	-0.45	-0.45	0.03	-0.39
rob_high	0.01	-0.01	-0.21	0.21	0.37	-0.37	-0.33	0.79	-0.23	-0.09	-0.21	-0.66	1	0.07	-0.07	-0.43	-0.34	0.22	0.8	-0.13	0.6
prereg	0.01	-0.01	-0.02	0.02	-0.22	0.22	0.34	0.31	-0.58	-0.22	0.36	-0.33	0.07	1	-1	0.3	0.04	-0.24	-0.13	0.32	-0.31
conventional	-0.01	0.01	0.02	-0.02	0.22	-0.22	-0.34	-0.31	0.58	0.22	-0.36	0.33	-0.07	-1	1	-0.3	-0.04	0.24	0.13	-0.32	0.31
age	-0.33	0.33	-0.3	0.3	-0.36	0.36	-0.24	0.02	0.27	0.02	-0.22	0.52	-0.43	0.3	-0.3	1	-0.04	-0.57	-0.08	0.19	-0.38
male	0.52	-0.52	0.07	-0.07	-0.51	0.51	0.23	-0.28	-0.21	0.37	0.34	0.02	-0.34	0.04	-0.04	-0.04	1	-0.19	-0.49	-0.02	-0.57
dur	-0.1	0.1	0.43	-0.43	0.64	-0.64	0.05	-0.06	0.01	-0.03	0.36	-0.45	0.22	-0.24	0.24	-0.57	-0.19	1	0.04	0.04	0.61
int	0.03	-0.03	-0.1	0.1	0.35	-0.35	-0.56	0.75	0.04	0.01	-0.27	-0.45	0.8	-0.13	0.13	-0.08	-0.49	0.04	1	-0.27	0.65
FU	-0.25	0.25	-0.06	0.06	-0.13	0.13	0.09	-0.02	-0.03	-0.13	0.1	0.03	-0.13	0.32	-0.32	0.19	-0.02	0.04	-0.27	1	-0.22
sess	-0.05	0.05	0.15	-0.15	0.93	-0.93	-0.17	0.23	0.02	-0.01	-0.14	-0.39	0.6	-0.31	0.31	-0.38	-0.57	0.61	0.65	-0.22	1

Social functioning (degree of impairment)

Excluded factors due to no appearance: waitlist-only

cor_mat_dat_cat_social <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Social") & test_type != "Raw events") |> 
  select(
    samp = schizophrenia,
    treat = CBT_int,
    test = test_type,
    str = analysis_strategy,
    des = QES_design,
    ctr = control,
    rob = overall_rob,
    pre = prereg_chr
  )

cat_dummy_dat_social <- 
  fastDummies::dummy_cols(cor_mat_dat_cat_social) %>% 
  select(where(is.numeric))


cor_mat_dat_con_social <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Social") & test_type != "Raw events") |> 
  select(
    age = age_mean,
    male = male_pct,
    dur = duration_in_weeks,
    int = sessions_per_week,
    FU = time_after_end_intervention_weeks,
    sess = total_number_of_sessions
  )

cor_mat_dat_social <- 
  bind_cols(cat_dummy_dat_social, cor_mat_dat_con_social) |> 
  select(
    s_schizo = samp_Schizophrenia,
    s_oth = samp_Other,
    cbt_trt = treat_CBT, 
    oth_trt = treat_Other,
    test_clin = `test_Clinician-rated measure`,
    test_self = `test_Self-reported`,
    itt = str_ITT,
    tot = str_TOT,
    qes = des_QES,
    rct = des_RCT,
    tau = ctr_TAU,
    tau_wait = `ctr_TAU with Waiting-list`,
    ind_trt = `ctr_Individual treatment` ,
    rob_low = rob_Low,
    rob_mod = `rob_Some concerns/Moderate`,
    rob_high = `rob_Serious/High`,
    prereg = pre_Preregistered, 
    conventional = `pre_Not preregistered`,
    everything()
  ) |> 
  na.omit()

cor_mat_social <- 
  cor(
    model.matrix(
      reformulate(names(cor_mat_dat_social[,1:ncol(cor_mat_dat_social)])), 
      data = cor_mat_dat_social
      )
  ) |> 
  as.data.frame() |>  
  select(-c(`(Intercept)`)) |> 
  na.omit() |> 
  mutate(
    across(.cols = everything(), ~ round(.x, 2))
  )

cor_mat_formatted_social <- 
  cor_mat_social |> 
  mutate(
    across(.cols = everything(), ~ cell_spec(.x, bold = ifelse(abs(.x) > 0.5, T, F)))
  )

kbl(
  cor_mat_formatted_social, 
  escape = F,
  col.names = c("Category", colnames(cor_mat_social))
  ) |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "100%", fixed_thead = TRUE)

Table 36: Correlation matrix based on social functioning outcomes only.

Category	s_schizo	s_oth	cbt_trt	oth_trt	test_clin	test_self	itt	tot	qes	rct	tau	tau_wait	ind_trt	rob_low	rob_mod	rob_high	prereg	conventional	age	male	dur	int	FU	sess
s_schizo	1	-1	-0.55	0.55	-0.11	0.11	0.44	-0.44	-0.09	0.09	0.11	-0.24	0.26	0.45	-0.34	-0.21	0.3	-0.3	-0.13	0.73	0.71	0.08	0.08	0.71
s_oth	-1	1	0.55	-0.55	0.11	-0.11	-0.44	0.44	0.09	-0.09	-0.11	0.24	-0.26	-0.45	0.34	0.21	-0.3	0.3	0.13	-0.73	-0.71	-0.08	-0.08	-0.71
cbt_trt	-0.55	0.55	1	-1	0.09	-0.09	-0.35	0.35	-0.14	0.14	0.03	0.03	-0.14	0.14	-0.01	-0.19	0.14	-0.14	0.24	-0.54	-0.6	-0.16	0.01	-0.6
oth_trt	0.55	-0.55	-1	1	-0.09	0.09	0.35	-0.35	0.14	-0.14	-0.03	-0.03	0.14	-0.14	0.01	0.19	-0.14	0.14	-0.24	0.54	0.6	0.16	-0.01	0.6
test_clin	-0.11	0.11	0.09	-0.09	1	-1	0.16	-0.16	0.35	-0.35	-0.15	0.19	-0.07	-0.18	0.17	0.04	-0.39	0.39	-0.37	-0.06	0.04	0.19	-0.06	0.15
test_self	0.11	-0.11	-0.09	0.09	-1	1	-0.16	0.16	-0.35	0.35	0.15	-0.19	0.07	0.18	-0.17	-0.04	0.39	-0.39	0.37	0.06	-0.04	-0.19	0.06	-0.15
itt	0.44	-0.44	-0.35	0.35	0.16	-0.16	1	-1	-0.17	0.17	-0.16	0.24	-0.17	-0.04	0.01	0.05	0.32	-0.32	-0.31	0.42	0.25	0.17	-0.16	0.2
tot	-0.44	0.44	0.35	-0.35	-0.16	0.16	-1	1	0.17	-0.17	0.16	-0.24	0.17	0.04	-0.01	-0.05	-0.32	0.32	0.31	-0.42	-0.25	-0.17	0.16	-0.2
qes	-0.09	0.09	-0.14	0.14	0.35	-0.35	-0.17	0.17	1	-1	0.07	-0.06	-0.02	-0.2	-0.09	0.43	-0.3	0.3	0.03	0.11	0.41	0	0.27	0.36
rct	0.09	-0.09	0.14	-0.14	-0.35	0.35	0.17	-0.17	-1	1	-0.07	0.06	0.02	0.2	0.09	-0.43	0.3	-0.3	-0.03	-0.11	-0.41	0	-0.27	-0.36
tau	0.11	-0.11	0.03	-0.03	-0.15	0.15	-0.16	0.16	0.07	-0.07	1	-0.91	-0.35	0.18	0.26	-0.63	-0.07	0.07	0.63	0.4	-0.03	-0.2	0.34	-0.09
tau_wait	-0.24	0.24	0.03	-0.03	0.19	-0.19	0.24	-0.24	-0.06	0.06	-0.91	1	-0.06	-0.25	-0.24	0.69	0.04	-0.04	-0.49	-0.53	-0.15	0.2	-0.31	-0.07
ind_trt	0.26	-0.26	-0.14	0.14	-0.07	0.07	-0.17	0.17	-0.02	0.02	-0.35	-0.06	1	0.12	-0.09	-0.06	0.08	-0.08	-0.41	0.24	0.41	0.02	-0.12	0.39
rob_low	0.45	-0.45	0.14	-0.14	-0.18	0.18	-0.04	0.04	-0.2	0.2	0.18	-0.25	0.12	1	-0.76	-0.47	0.67	-0.67	0.16	0.11	0.35	-0.62	0.19	0.1
rob_mod	-0.34	0.34	-0.01	0.01	0.17	-0.17	0.01	-0.01	-0.09	0.09	0.26	-0.24	-0.09	-0.76	1	-0.21	-0.62	0.62	0.12	0.11	-0.44	0.54	-0.14	-0.19
rob_high	-0.21	0.21	-0.19	0.19	0.04	-0.04	0.05	-0.05	0.43	-0.43	-0.63	0.69	-0.06	-0.47	-0.21	1	-0.17	0.17	-0.4	-0.32	0.07	0.2	-0.1	0.11
prereg	0.3	-0.3	0.14	-0.14	-0.39	0.39	0.32	-0.32	-0.3	0.3	-0.07	0.04	0.08	0.67	-0.62	-0.17	1	-1	0.17	0.02	0.14	-0.43	-0.05	-0.07
conventional	-0.3	0.3	-0.14	0.14	0.39	-0.39	-0.32	0.32	0.3	-0.3	0.07	-0.04	-0.08	-0.67	0.62	0.17	-1	1	-0.17	-0.02	-0.14	0.43	0.05	0.07
age	-0.13	0.13	0.24	-0.24	-0.37	0.37	-0.31	0.31	0.03	-0.03	0.63	-0.49	-0.41	0.16	0.12	-0.4	0.17	-0.17	1	0.04	-0.23	-0.07	0.16	-0.22
male	0.73	-0.73	-0.54	0.54	-0.06	0.06	0.42	-0.42	0.11	-0.11	0.4	-0.53	0.24	0.11	0.11	-0.32	0.02	-0.02	0.04	1	0.66	0.3	0.11	0.63
dur	0.71	-0.71	-0.6	0.6	0.04	-0.04	0.25	-0.25	0.41	-0.41	-0.03	-0.15	0.41	0.35	-0.44	0.07	0.14	-0.14	-0.23	0.66	1	-0.05	0.2	0.88
int	0.08	-0.08	-0.16	0.16	0.19	-0.19	0.17	-0.17	0	0	-0.2	0.2	0.02	-0.62	0.54	0.2	-0.43	0.43	-0.07	0.3	-0.05	1	-0.27	0.35
FU	0.08	-0.08	0.01	-0.01	-0.06	0.06	-0.16	0.16	0.27	-0.27	0.34	-0.31	-0.12	0.19	-0.14	-0.1	-0.05	0.05	0.16	0.11	0.2	-0.27	1	0.08
sess	0.71	-0.71	-0.6	0.6	0.15	-0.15	0.2	-0.2	0.36	-0.36	-0.09	-0.07	0.39	0.1	-0.19	0.11	-0.07	0.07	-0.22	0.63	0.88	0.35	0.08	1

Wellbeing and quality of life

Excluded factors due to no appearance: waitlist-only

cor_mat_dat_cat_well <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Well") & test_type != "Raw events") |> 
  select(
    samp = schizophrenia,
    treat = CBT_int,
    test = test_type,
    str = analysis_strategy,
    des = QES_design,
    ctr = control,
    rob = overall_rob,
    pre = prereg_chr
  )

cat_dummy_dat_well <- 
  fastDummies::dummy_cols(cor_mat_dat_cat_well) %>% 
  select(where(is.numeric))


cor_mat_dat_con_well <- 
  reintegration_dat |> 
  filter(str_detect(analysis_plan, "Well") & test_type != "Raw events") |> 
  select(
    age = age_mean,
    male = male_pct,
    dur = duration_in_weeks,
    int = sessions_per_week,
    FU = time_after_end_intervention_weeks,
    sess = total_number_of_sessions
  )

cor_mat_dat_well <- 
  bind_cols(cat_dummy_dat_well, cor_mat_dat_con_well) |> 
  select(
    s_schizo = samp_Schizophrenia,
    s_oth = samp_Other,
    cbt_trt = treat_CBT, 
    oth_trt = treat_Other,
    test_clin = `test_Clinician-rated measure`,
    test_self = `test_Self-reported`,
    itt = str_ITT,
    tot = str_TOT,
    qes = des_QES,
    rct = des_RCT,
    tau = ctr_TAU,
    tau_wait = `ctr_TAU with Waiting-list`,
    wait = `ctr_Waiting-list only`,
    ind_trt = `ctr_Individual treatment` ,
    rob_low = rob_Low,
    rob_mod = `rob_Some concerns/Moderate`,
    rob_high = `rob_Serious/High`,
    prereg = pre_Preregistered, 
    conventional = `pre_Not preregistered`,
    everything()
  ) |> 
  na.omit()

cor_mat_well <- 
  cor(
    model.matrix(
      reformulate(names(cor_mat_dat_well[,1:ncol(cor_mat_dat_well)])), 
      data = cor_mat_dat_well
      )
  ) |> 
  as.data.frame() |>  
  select(-c(`(Intercept)`)) |> 
  na.omit() |> 
  mutate(
    across(.cols = everything(), ~ round(.x, 2))
  )

cor_mat_formatted_well <- 
  cor_mat_well |> 
  mutate(
    across(.cols = everything(), ~ cell_spec(.x, bold = ifelse(abs(.x) > 0.5, T, F)))
  )

kbl(
  cor_mat_formatted_well, 
  escape = F,
  col.names = c("Category", colnames(cor_mat_well))
  ) |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "100%", fixed_thead = TRUE)

Table 37: Correlation matrix based on wellbeing and quality of life outcomes only.

Category	s_schizo	s_oth	cbt_trt	oth_trt	test_clin	test_self	itt	tot	qes	rct	tau	tau_wait	wait	ind_trt	rob_low	rob_mod	rob_high	prereg	conventional	age	male	dur	int	FU	sess
s_schizo	1	-1	-0.32	0.32	0.17	-0.17	0.22	-0.22	0.07	-0.07	0.18	-0.15	-0.07	-0.05	0.19	-0.28	0.1	-0.03	0.03	-0.36	0.52	0.54	-0.02	0.07	0.51
s_oth	-1	1	0.32	-0.32	-0.17	0.17	-0.22	0.22	-0.07	0.07	-0.18	0.15	0.07	0.05	-0.19	0.28	-0.1	0.03	-0.03	0.36	-0.52	-0.54	0.02	-0.07	-0.51
cbt_trt	-0.32	0.32	1	-1	-0.09	0.09	-0.28	0.28	-0.01	0.01	0.16	-0.34	0.21	0.15	0.17	-0.04	-0.19	0.14	-0.14	-0.01	-0.72	-0.24	-0.01	0.27	-0.26
oth_trt	0.32	-0.32	-1	1	0.09	-0.09	0.28	-0.28	0.01	-0.01	-0.16	0.34	-0.21	-0.15	-0.17	0.04	0.19	-0.14	0.14	0.01	0.72	0.24	0.01	-0.27	0.26
test_clin	0.17	-0.17	-0.09	0.09	1	-1	-0.12	0.12	0.27	-0.27	0.17	-0.14	-0.06	-0.04	-0.34	0.05	0.4	-0.42	0.42	-0.25	0.24	-0.15	0.1	-0.15	0.04
test_self	-0.17	0.17	0.09	-0.09	-1	1	0.12	-0.12	-0.27	0.27	-0.17	0.14	0.06	0.04	0.34	-0.05	-0.4	0.42	-0.42	0.25	-0.24	0.15	-0.1	0.15	-0.04
itt	0.22	-0.22	-0.28	0.28	-0.12	0.12	1	-1	-0.27	0.27	0.44	-0.37	-0.16	-0.12	-0.42	0.54	-0.14	0.03	-0.03	0.03	0.26	0.22	-0.09	-0.05	0.11
tot	-0.22	0.22	0.28	-0.28	0.12	-0.12	-1	1	0.27	-0.27	-0.44	0.37	0.16	0.12	0.42	-0.54	0.14	-0.03	0.03	-0.03	-0.26	-0.22	0.09	0.05	-0.11
qes	0.07	-0.07	-0.01	0.01	0.27	-0.27	-0.27	0.27	1	-1	-0.16	-0.12	0.28	0.43	-0.28	-0.09	0.52	-0.55	0.55	-0.11	0.07	-0.14	0.47	-0.17	0.21
rct	-0.07	0.07	0.01	-0.01	-0.27	0.27	0.27	-0.27	-1	1	0.16	0.12	-0.28	-0.43	0.28	0.09	-0.52	0.55	-0.55	0.11	-0.07	0.14	-0.47	0.17	-0.21
tau	0.18	-0.18	0.16	-0.16	0.17	-0.17	0.44	-0.44	-0.16	0.16	1	-0.84	-0.37	-0.26	-0.09	0.19	-0.13	-0.06	0.06	-0.05	-0.17	0.18	-0.34	0.26	-0.13
tau_wait	-0.15	0.15	-0.34	0.34	-0.14	0.14	-0.37	0.37	-0.12	0.12	-0.84	1	-0.07	-0.05	0.23	-0.3	0.08	0.21	-0.21	-0.05	0.26	-0.13	0.05	-0.22	0
wait	-0.07	0.07	0.21	-0.21	-0.06	0.06	-0.16	0.16	0.28	-0.28	-0.37	-0.07	1	-0.02	-0.18	0.24	-0.08	-0.12	0.12	0.31	-0.09	-0.09	0.67	-0.11	0.32
ind_trt	-0.05	0.05	0.15	-0.15	-0.04	0.04	-0.12	0.12	0.43	-0.43	-0.26	-0.05	-0.02	1	-0.12	-0.09	0.29	-0.24	0.24	-0.13	-0.05	-0.05	-0.01	-0.05	-0.04
rob_low	0.19	-0.19	0.17	-0.17	-0.34	0.34	-0.42	0.42	-0.28	0.28	-0.09	0.23	-0.18	-0.12	1	-0.73	-0.42	0.52	-0.52	-0.18	-0.13	0.41	-0.31	0.33	-0.01
rob_mod	-0.28	0.28	-0.04	0.04	0.05	-0.05	0.54	-0.54	-0.09	0.09	0.19	-0.3	0.24	-0.09	-0.73	1	-0.32	-0.32	0.32	0.33	0.09	-0.27	0.26	-0.27	0.03
rob_high	0.1	-0.1	-0.19	0.19	0.4	-0.4	-0.14	0.14	0.52	-0.52	-0.13	0.08	-0.08	0.29	-0.42	-0.32	1	-0.3	0.3	-0.18	0.06	-0.21	0.08	-0.1	-0.03
prereg	-0.03	0.03	0.14	-0.14	-0.42	0.42	0.03	-0.03	-0.55	0.55	-0.06	0.21	-0.12	-0.24	0.52	-0.32	-0.3	1	-1	0.36	-0.37	0.05	-0.35	0.2	-0.27
conventional	0.03	-0.03	-0.14	0.14	0.42	-0.42	-0.03	0.03	0.55	-0.55	0.06	-0.21	0.12	0.24	-0.52	0.32	0.3	-1	1	-0.36	0.37	-0.05	0.35	-0.2	0.27
age	-0.36	0.36	-0.01	0.01	-0.25	0.25	0.03	-0.03	-0.11	0.11	-0.05	-0.05	0.31	-0.13	-0.18	0.33	-0.18	0.36	-0.36	1	-0.19	-0.28	0.03	-0.12	-0.14
male	0.52	-0.52	-0.72	0.72	0.24	-0.24	0.26	-0.26	0.07	-0.07	-0.17	0.26	-0.09	-0.05	-0.13	0.09	0.06	-0.37	0.37	-0.19	1	0.39	-0.02	-0.28	0.38
dur	0.54	-0.54	-0.24	0.24	-0.15	0.15	0.22	-0.22	-0.14	0.14	0.18	-0.13	-0.09	-0.05	0.41	-0.27	-0.21	0.05	-0.05	-0.28	0.39	1	-0.15	0.26	0.62
int	-0.02	0.02	-0.01	0.01	0.1	-0.1	-0.09	0.09	0.47	-0.47	-0.34	0.05	0.67	-0.01	-0.31	0.26	0.08	-0.35	0.35	0.03	-0.02	-0.15	1	-0.18	0.56
FU	0.07	-0.07	0.27	-0.27	-0.15	0.15	-0.05	0.05	-0.17	0.17	0.26	-0.22	-0.11	-0.05	0.33	-0.27	-0.1	0.2	-0.2	-0.12	-0.28	0.26	-0.18	1	0.03
sess	0.51	-0.51	-0.26	0.26	0.04	-0.04	0.11	-0.11	0.21	-0.21	-0.13	0	0.32	-0.04	-0.01	0.03	-0.03	-0.27	0.27	-0.14	0.38	0.62	0.56	0.03	1

Mental health outcomes

cor_mat_dat_cat_mental <- 
  mental_health_dat |> 
  filter(test_type != "Raw events") |> 
  select(
    plan = analysis_plan,
    samp = schizophrenia,
    treat = CBT_int,
    test = test_type,
    str = analysis_strategy,
    des = QES_design,
    ctr = control,
    rob = overall_rob,
    pre = prereg_chr
  )

cat_dummy_dat_mental <- 
  fastDummies::dummy_cols(cor_mat_dat_cat_mental) %>% 
  select(where(is.numeric))


cor_mat_dat_con_mental <- 
  mental_health_dat |>  
  filter(test_type != "Raw events") |> 
  select(
    age = age_mean,
    male = male_pct,
    dur = duration_in_weeks,
    int = sessions_per_week,
    FU = time_after_end_intervention_weeks,
    sess = total_number_of_sessions
  )

cor_mat_dat_mental <- 
  bind_cols(cat_dummy_dat_mental, cor_mat_dat_con_mental) |> 
  select(
    anxiety = plan_Anxiety,
    depress = plan_Depression,
    g_mental = `plan_General mental health`,
    symptoms = `plan_Symptoms of psychosis`,
    s_schizo = samp_Schizophrenia,
    s_oth = samp_Other,
    cbt_trt = treat_CBT, 
    oth_trt = treat_Other,
    test_clin = `test_Clinician-rated measure`,
    test_self = `test_Self-reported`,
    itt = str_ITT,
    tot = str_TOT,
    qes = des_QES,
    rct = des_RCT,
    tau = ctr_TAU,
    tau_wait = `ctr_TAU with Waiting-list`,
    wait = `ctr_Waiting-list only`,
    ind_trt = `ctr_Individual treatment` ,
    rob_low = rob_Low,
    rob_mod = `rob_Some concerns/Moderate`,
    rob_high = `rob_Serious/High`,
    prereg = pre_Preregistered, 
    conventional = `pre_Not preregistered`,
    everything()
  ) |> 
  na.omit()

cor_mat_mental <- 
  cor(
    model.matrix(
      reformulate(names(cor_mat_dat_mental[,1:ncol(cor_mat_dat_mental)])), 
      data = cor_mat_dat_mental
      )
  ) %>% 
  as.data.frame() %>% 
  select(-c(`(Intercept)`)) %>% 
  na.omit() |> 
  mutate(
    
    across(.cols = everything(), ~ round(.x, 2))
  )

cor_mat_formatted_mental <- 
  cor_mat_mental |> 
  mutate(
    across(.cols = everything(), ~ cell_spec(.x, bold = ifelse(abs(.x) > 0.5, T, F)))
  )


kbl(
  cor_mat_formatted_mental, 
  escape = F,
  col.names = c("Category", colnames(cor_mat_mental))
  ) |> 
  kable_styling(
    bootstrap_options = c("striped", "hover"),
    font_size = 10
  ) |> 
  scroll_box(width = "100%", height = "100%", fixed_thead = TRUE)

Table 38: Correlation matrix across all mental health outcomes.

Category	anxiety	depress	g_mental	symptoms	s_schizo	s_oth	cbt_trt	oth_trt	test_clin	test_self	itt	tot	qes	rct	tau	tau_wait	wait	ind_trt	rob_low	rob_mod	rob_high	prereg	conventional	age	male	dur	int	FU	sess
anxiety	1	-0.2	-0.33	-0.14	-0.09	0.09	0.32	-0.32	-0.11	0.11	-0.14	0.14	-0.05	0.05	-0.16	-0.01	0.25	0.11	0.08	-0.09	0.02	0.09	-0.09	0.22	-0.03	-0.06	-0.07	-0.03	-0.13
depress	-0.2	1	-0.58	-0.25	-0.16	0.16	0.04	-0.04	-0.13	0.13	-0.02	0.02	-0.02	0.02	0.09	-0.17	0.09	0.03	-0.15	0.19	-0.07	-0.03	0.03	0.13	-0.11	-0.11	0	0.01	-0.1
g_mental	-0.33	-0.58	1	-0.41	-0.11	0.11	-0.08	0.08	-0.1	0.1	0.19	-0.19	-0.15	0.15	-0.15	0.29	-0.16	-0.05	0.1	-0.03	-0.06	0.19	-0.19	-0.27	-0.12	0.01	-0.01	-0.01	0.01
symptoms	-0.14	-0.25	-0.41	1	0.43	-0.43	-0.21	0.21	0.39	-0.39	-0.13	0.13	0.27	-0.27	0.23	-0.18	-0.1	-0.06	-0.01	-0.11	0.16	-0.29	0.29	0.03	0.33	0.17	0.08	0.03	0.22
s_schizo	-0.09	-0.16	-0.11	0.43	1	-1	-0.2	0.2	0	0	0.12	-0.12	0.07	-0.07	0.09	-0.12	-0.06	0.15	0.35	-0.31	0.01	0.12	-0.12	-0.18	0.34	0.46	-0.06	0.04	0.31
s_oth	0.09	0.16	0.11	-0.43	-1	1	0.2	-0.2	0	0	-0.12	0.12	-0.07	0.07	-0.09	0.12	0.06	-0.15	-0.35	0.31	-0.01	-0.12	0.12	0.18	-0.34	-0.46	0.06	-0.04	-0.31
cbt_trt	0.32	0.04	-0.08	-0.21	-0.2	0.2	1	-1	0.01	-0.01	-0.02	0.02	-0.08	0.08	-0.12	-0.1	0.33	0.11	0.09	-0.03	-0.06	0.13	-0.13	0.17	-0.08	-0.32	0.15	0.13	-0.24
oth_trt	-0.32	-0.04	0.08	0.21	0.2	-0.2	-1	1	-0.01	0.01	0.02	-0.02	0.08	-0.08	0.12	0.1	-0.33	-0.11	-0.09	0.03	0.06	-0.13	0.13	-0.17	0.08	0.32	-0.15	-0.13	0.24
test_clin	-0.11	-0.13	-0.1	0.39	0	0	0.01	-0.01	1	-1	0.03	-0.03	0.05	-0.05	0.24	-0.28	0.07	-0.1	-0.27	0.22	0.01	-0.25	0.25	0.15	0.08	-0.07	0.04	-0.01	0.03
test_self	0.11	0.13	0.1	-0.39	0	0	-0.01	0.01	-1	1	-0.03	0.03	-0.05	0.05	-0.24	0.28	-0.07	0.1	0.27	-0.22	-0.01	0.25	-0.25	-0.15	-0.08	0.07	-0.04	0.01	-0.03
itt	-0.14	-0.02	0.19	-0.13	0.12	-0.12	-0.02	0.02	0.03	-0.03	1	-1	-0.19	0.19	0.33	-0.19	-0.23	-0.15	-0.2	0.26	-0.1	0.27	-0.27	-0.32	-0.22	0.16	-0.14	0.03	0.03
tot	0.14	0.02	-0.19	0.13	-0.12	0.12	0.02	-0.02	-0.03	0.03	-1	1	0.19	-0.19	-0.33	0.19	0.23	0.15	0.2	-0.26	0.1	-0.27	0.27	0.32	0.22	-0.16	0.14	-0.03	-0.03
qes	-0.05	-0.02	-0.15	0.27	0.07	-0.07	-0.08	0.08	0.05	-0.05	-0.19	0.19	1	-1	0	-0.16	0.11	0.25	-0.22	-0.33	0.67	-0.44	0.44	-0.08	0.03	0.15	0.39	-0.13	0.42
rct	0.05	0.02	0.15	-0.27	-0.07	0.07	0.08	-0.08	-0.05	0.05	0.19	-0.19	-1	1	0	0.16	-0.11	-0.25	0.22	0.33	-0.67	0.44	-0.44	0.08	-0.03	-0.15	-0.39	0.13	-0.42
tau	-0.16	0.09	-0.15	0.23	0.09	-0.09	-0.12	0.12	0.24	-0.24	0.33	-0.33	0	0	1	-0.79	-0.42	-0.27	-0.16	0.31	-0.22	-0.08	0.08	-0.07	-0.05	0.08	-0.21	0.24	-0.04
tau_wait	-0.01	-0.17	0.29	-0.18	-0.12	0.12	-0.1	0.1	-0.28	0.28	-0.19	0.19	-0.16	0.16	-0.79	1	-0.1	-0.06	0.25	-0.41	0.25	0.17	-0.17	-0.2	0.03	-0.11	-0.02	-0.16	-0.08
wait	0.25	0.09	-0.16	-0.1	-0.06	0.06	0.33	-0.33	0.07	-0.07	-0.23	0.23	0.11	-0.11	-0.42	-0.1	1	-0.03	-0.13	0.2	-0.11	-0.07	0.07	0.55	0	-0.1	0.46	-0.14	0.12
ind_trt	0.11	0.03	-0.05	-0.06	0.15	-0.15	0.11	-0.11	-0.1	0.1	-0.15	0.15	0.25	-0.25	-0.27	-0.06	-0.03	1	0.03	-0.17	0.18	-0.08	0.08	-0.14	0.06	0.2	-0.03	-0.06	0.13
rob_low	0.08	-0.15	0.1	-0.01	0.35	-0.35	0.09	-0.09	-0.27	0.27	-0.2	0.2	-0.22	0.22	-0.16	0.25	-0.13	0.03	1	-0.66	-0.28	0.49	-0.49	0.02	0.1	0.3	-0.24	0.19	-0.02
rob_mod	-0.09	0.19	-0.03	-0.11	-0.31	0.31	-0.03	0.03	0.22	-0.22	0.26	-0.26	-0.33	0.33	0.31	-0.41	0.2	-0.17	-0.66	1	-0.54	-0.22	0.22	0.13	-0.05	-0.31	0.16	-0.05	-0.15
rob_high	0.02	-0.07	-0.06	0.16	0.01	-0.01	-0.06	0.06	0.01	-0.01	-0.1	0.1	0.67	-0.67	-0.22	0.25	-0.11	0.18	-0.28	-0.54	1	-0.27	0.27	-0.19	-0.05	0.07	0.07	-0.14	0.22
prereg	0.09	-0.03	0.19	-0.29	0.12	-0.12	0.13	-0.13	-0.25	0.25	0.27	-0.27	-0.44	0.44	-0.08	0.17	-0.07	-0.08	0.49	-0.22	-0.27	1	-1	0.2	-0.22	0.01	-0.26	0.1	-0.24
conventional	-0.09	0.03	-0.19	0.29	-0.12	0.12	-0.13	0.13	0.25	-0.25	-0.27	0.27	0.44	-0.44	0.08	-0.17	0.07	0.08	-0.49	0.22	0.27	-1	1	-0.2	0.22	-0.01	0.26	-0.1	0.24
age	0.22	0.13	-0.27	0.03	-0.18	0.18	0.17	-0.17	0.15	-0.15	-0.32	0.32	-0.08	0.08	-0.07	-0.2	0.55	-0.14	0.02	0.13	-0.19	0.2	-0.2	1	0.05	-0.21	0.1	-0.09	-0.1
male	-0.03	-0.11	-0.12	0.33	0.34	-0.34	-0.08	0.08	0.08	-0.08	-0.22	0.22	0.03	-0.03	-0.05	0.03	0	0.06	0.1	-0.05	-0.05	-0.22	0.22	0.05	1	0.1	0	-0.14	0
dur	-0.06	-0.11	0.01	0.17	0.46	-0.46	-0.32	0.32	-0.07	0.07	0.16	-0.16	0.15	-0.15	0.08	-0.11	-0.1	0.2	0.3	-0.31	0.07	0.01	-0.01	-0.21	0.1	1	-0.15	0.18	0.72
int	-0.07	0	-0.01	0.08	-0.06	0.06	0.15	-0.15	0.04	-0.04	-0.14	0.14	0.39	-0.39	-0.21	-0.02	0.46	-0.03	-0.24	0.16	0.07	-0.26	0.26	0.1	0	-0.15	1	-0.17	0.43
FU	-0.03	0.01	-0.01	0.03	0.04	-0.04	0.13	-0.13	-0.01	0.01	0.03	-0.03	-0.13	0.13	0.24	-0.16	-0.14	-0.06	0.19	-0.05	-0.14	0.1	-0.1	-0.09	-0.14	0.18	-0.17	1	-0.02
sess	-0.13	-0.1	0.01	0.22	0.31	-0.31	-0.24	0.24	0.03	-0.03	0.03	-0.03	0.42	-0.42	-0.04	-0.08	0.12	0.13	-0.02	-0.15	0.22	-0.24	0.24	-0.1	0	0.72	0.43	-0.02	1

Standard Errors and Other Auxiliary Data

SE forest plot

This forest plot shows the within- and between study variation of standard error of effect size estimates for reintegrational outcomes.

reintegration_dat |> 
  arrange(desc(study)) |>
  mutate(
    study = factor(study, unique(study)),
    segt_pop = sqrt(vgt_pop)
    ) |> 
  ggplot(aes(x = segt_pop, y = study, color = study)) +
  geom_point(aes(size = N_total), alpha = 0.5) +
  scale_y_discrete() +
  theme_minimal() +
  labs(y = "", x = "SE") +
  theme(legend.position = "none", axis.text.y = element_text(size = 10))

Figure 83: Forest plot of standard error of reintegrational effect size estimates.

mental_health_dat |> 
  arrange(desc(study)) |>
  mutate(
    study = factor(study, unique(study)),
    segt_pop = sqrt(vgt_pop)
    ) |> 
  ggplot(aes(x = segt_pop, y = study, color = study)) +
  geom_point(aes(size = N_total), alpha = 0.5) +
  scale_y_discrete() +
  theme_minimal() +
  labs(y = "", x = "SE") +
  theme(legend.position = "none", axis.text.y = element_text(size = 10))

Figure 84: Forest plot of standard error of mental health effect size estimates.

Modified SE forest plot

reintegration_dat |> 
  arrange(desc(study)) |>
  mutate(
    study = factor(study, unique(study)),
    Wse_pop = sqrt(Wgt_pop)
    ) |> 
  ggplot(aes(x = Wse_pop, y = study, color = study)) +
  geom_point(aes(size = N_total), alpha = 0.5, position = position_jitter(height = 0.4)) +
  scale_y_discrete() +
  theme_minimal() +
  labs(y = "", x = "Modified SE") +
  theme(legend.position = "none", axis.text.y = element_text(size = 10))

Figure 85: Forest plot of modified standard error of reintegrational effect size estimates.

mental_health_dat |> 
  arrange(desc(study)) |>
  mutate(
    study = factor(study, unique(study)),
    Wse_pop = sqrt(Wgt_pop)
    ) |> 
  ggplot(aes(x = Wse_pop, y = study, color = study)) +
  geom_point(aes(size = N_total), alpha = 0.5, position = position_jitter(height = 0.4)) +
  scale_y_discrete() +
  theme_minimal() +
  labs(y = "", x = "Modified SE") +
  theme(legend.position = "none", axis.text.y = element_text(size = 10))

Figure 86: Forest plot of modified standard error of mental health effect size estimates.

Inverse sampling covariance weights plot

iscw <-
  function(k, rho, v) {
    iscw_weights <-  k / (((k - 1) * rho + 1) * v)
    return(iscw_weights)
  }

rho_val <- round(seq(0, 0.8, 0.2), 1)

ISCW_plot <- 
  reintegration_dat |> 
  expand_grid(rho = rho_val) |> 
  group_by(study, rho) |> 
  summarise(
    k = n(),
    v_bar = mean(vgt_pop)
  ) |> 
  ungroup() |> 
  mutate(iscw_w = iscw(k, rho, v_bar),
         depd = ifelse(k > 1, 1, 0)) |> 
  group_by(rho) |> 
  mutate(iscw_w_norm = iscw_w / sum(iscw_w)) |> 
  ungroup() |> 
  distinct(study, rho, iscw_w_norm, depd) |> 
  mutate(study = factor(study, levels = study[rho == 0.8][order(iscw_w_norm[rho == 0.8])])) |> 
  ggplot(aes(y = iscw_w_norm, x = study, group = factor(rho), colour = factor(rho))) +
  geom_point() +
  geom_line() + 
  theme_minimal() +
  coord_flip() +
  theme(
    legend.position = "inside",
    legend.position.inside = c(0.8, 0.2)
  ) +
  labs(x = "", y = "Normalized Weight", colour = "Assumed Correlation") 

ISCW_plot

Figure 87: Plot of inverse sampling covariance (ISC) weights for each study (reintegrational outcomes)

ISCW_plot_mental <- 
  mental_health_dat |> 
  expand_grid(rho = rho_val) |> 
  group_by(study, rho) |> 
  summarise(
    k = n(),
    v_bar = mean(vgt_pop)
  ) |> 
  ungroup() |> 
  mutate(iscw_w = iscw(k, rho, v_bar),
         depd = ifelse(k > 1, 1, 0)) |> 
  group_by(rho) |> 
  mutate(iscw_w_norm = iscw_w / sum(iscw_w)) |> 
  ungroup() |> 
  distinct(study, rho, iscw_w_norm, depd) |> 
  mutate(study = factor(study, levels = study[rho == 0.8][order(iscw_w_norm[rho == 0.8])])) |> 
  ggplot(aes(y = iscw_w_norm, x = study, group = factor(rho), colour = factor(rho))) +
  geom_point() +
  geom_line() + 
  theme_minimal() +
  coord_flip() +
  theme(
    legend.position = "inside",
    legend.position.inside = c(0.85, 0.2)
  ) +
  labs(x = "", y = "Normalized Weight", colour = "Assumed Correlation") 

ISCW_plot_mental

Figure 88: Plot of inverse sampling covariance (ISC) weights for each study (mental healt outcomes)

Study sample sizes versus standard error estimates

reintegration_dat |> 
  mutate(segt_pop = sqrt(vgt_pop)) |> 
  ggplot(aes(N_total, segt_pop)) + 
  geom_point() +
  geom_smooth() +
  theme_bw() +
  labs(x = "Total sample size", y = "Modified standard error")

Figure 89: Study sample sizes versus reintegrational standard error estimates

mental_health_dat |> 
  mutate(segt_pop = sqrt(vgt_pop)) |> 
  ggplot(aes(N_total, segt_pop)) + 
  geom_point() +
  geom_smooth() +
  theme_bw() +
  labs(x = "Total sample size", y = "Modified standard error")

Figure 90: Study sample sizes versus mental health standard error estimates

Standard error vs. scaled standrad error

reintegration_dat |> 
  mutate(
    sample_above_100 = if_else(N_total >100, "Sample > 100", "Sample < 100"),
    sample_above_100 = factor(sample_above_100, levels = c( "Sample > 100", "Sample < 100")),
    segt_pop = sqrt(vgt_pop),
    diff_se = sqrt(Wgt_pop) - sqrt(vgt_pop)
  ) |> 
  ggplot(aes(gt_pop, diff_se, color = sample_above_100)) + 
  geom_point() +
  geom_hline(yintercept = 0, linetype = "dashed") +
  theme_bw() +
  theme(
    plot.caption=element_text(hjust = 0, size = 10),
    legend.position= "bottom",
    legend.title = element_blank(),
    panel.spacing.x = unit(5, "mm"), 
    panel.spacing.y = unit(5, "mm"),
    #panel.grid.major = element_blank(),
    #panel.grid.minor = element_blank()
  ) +
  scale_x_continuous(expand=c(0,0), breaks = seq(-1L, 1.5, 0.2)) + 
  scale_y_continuous(expand=c(0,0), breaks = seq(-0.1, 0.1, 0.01)) + 
  expand_limits(x = c(-0.6, 1.5), y = c(-0.041, 0.01)) +
  labs(y = "Modified SE - SE (used in main analysis)", x = "Effect size estimate")

Figure 91: Standard error vs. modified standard errors (reintegration)

mental_health_dat |> 
  mutate(
    sample_above_100 = if_else(N_total >100, "Sample > 100", "Sample < 100"),
    sample_above_100 = factor(sample_above_100, levels = c( "Sample > 100", "Sample < 100")),
    segt_pop = sqrt(vgt_pop),
    diff_se = sqrt(Wgt_pop) - sqrt(vgt_pop)
  ) |> 
  ggplot(aes(gt_pop, diff_se, color = sample_above_100)) + 
  geom_point() +
  geom_hline(yintercept = 0, linetype = "dashed") +
  theme_bw() +
  theme(
    plot.caption=element_text(hjust = 0, size = 10),
    legend.position= "bottom",
    legend.title = element_blank(),
    panel.spacing.x = unit(5, "mm"), 
    panel.spacing.y = unit(5, "mm"),
    #panel.grid.major = element_blank(),
    #panel.grid.minor = element_blank()
  ) +
  scale_x_continuous(expand=c(0,0), breaks = seq(-1L, 2, 0.2)) + 
  scale_y_continuous(expand=c(0,0), breaks = seq(-0.1, 0.1, 0.01)) + 
  expand_limits(x = c(-0.71, 2), y = c(-0.091, 0.01)) +
  labs(y = "Modified SE - SE (used in main analysis)", x = "Effect size estimate")

Figure 92: Standard error vs. modified standard errors (mental health outcomes)

Effect size estimates distributions and outliers

Marginal distributions

reintegration_dat$gt_pop |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) |>
  knitr::kable(
    digits = 2,
    caption = "Marginal distribution of effect size estimates",
    booktabs = TRUE
  )  |> 
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Marginal distribution of effect size estimates

mean	sd	p0	p25	p50	p75	p100
0.21	0.28	-0.5	0.03	0.18	0.38	1.35

mental_health_dat$gt_pop |> 
  skim() |>
  select(-skim_type, -skim_variable, -n_missing, -complete_rate, -numeric.hist) |>
  rename_at(vars(starts_with("numeric.")), ~ str_remove(., "numeric\\.")) |>
  knitr::kable(
    digits = 2,
    caption = "Marginal distribution of effect size estimates",
    booktabs = TRUE
  )  |> 
  kable_styling(bootstrap_options = c("striped","condensed"), full_width = FALSE)

Marginal distribution of effect size estimates

mean	sd	p0	p25	p50	p75	p100
0.28	0.39	-0.64	0.02	0.21	0.53	1.8

Marginal distribution plots

qrtls <- quantile(reintegration_dat$gt_pop, c(.25, .75), na.rm = TRUE)
fences <-  qrtls + 3 * diff(qrtls) * c(-1, 1)
fence_dat <- data.frame(qrtl = qrtls, fence = fences)

es_dist_plot <- 
  ggplot(reintegration_dat, aes(gt_pop)) + 
  geom_density(fill = "cornflowerblue", alpha = 0.8) + 
  geom_vline(data = fence_dat, aes(xintercept = qrtl), linetype = "solid") + 
  geom_vline(data = fence_dat, aes(xintercept = fence), linetype = "dashed") + 
  geom_rug(alpha = 0.25) + 
  theme_minimal() + 
  theme(axis.title.y = element_blank()) +
  labs(x = "Effect size estimate")

es_dist_plot

Figure 93: Empirical distribution of reintegrational effect size estimates. Solid vertical lines indicate lower and upper quartiles. Dashed lines indicate the 1st quartile minus 3 times the inter-quartile range and the 3rd quartile plus 3 times the interquartile range. Effect sizes outside of the range of dashed lines would be considered outliers according to Tukey’s (1977) definition.

qrtls_mental <- quantile(mental_health_dat$gt_pop, c(.25, .75), na.rm = TRUE)
fences_mental <-  qrtls_mental + 3 * diff(qrtls_mental) * c(-1, 1)
fence_dat_mental <- data.frame(qrtl = qrtls_mental, fence = fences_mental)

es_dist_plot_mental <- 
  ggplot(mental_health_dat, aes(gt_pop)) + 
  geom_density(fill = "gray", alpha = 0.8) + 
  geom_vline(data = fence_dat_mental, aes(xintercept = qrtl), linetype = "solid") + 
  geom_vline(data = fence_dat_mental, aes(xintercept = fence), linetype = "dashed") + 
  geom_rug(alpha = 0.25) + 
  theme_minimal() + 
  theme(axis.title.y = element_blank()) +
  labs(x = "Effect size estimate")

es_dist_plot_mental

Figure 94: Empirical distribution of mental health effect size estimates. Solid vertical lines indicate lower and upper quartiles. Dashed lines indicate the 1st quartile minus 3 times the inter-quartile range and the 3rd quartile plus 3 times the interquartile range. Effect sizes outside of the range of dashed lines would be considered outliers according to Tukey’s (1977) definition.

Effect size distribution across type of outcomes

Reintegrational outcomes

outcomes_reint_dat <- 
  reintegration_dat |> 
  filter(!str_detect(analysis_plan, "Psychiatric")) |> 
  group_by(analysis_plan) |> 
  reframe(
    qrtl = quantile(gt,  c(.25, .75)),
    fence = qrtl + 3 * diff(qrtl) * c(-1,1),
    .groups = "drop"
  )


reintegration_outcome_plot <- 
  reintegration_dat |> 
  #Has only one data point
  filter(!str_detect(analysis_plan, "Psychiatric")) |> 
  ggplot(aes(x = gt_pop, fill = analysis_plan)) + 
  geom_density(alpha = 0.7) +
  geom_vline(data = outcomes_reint_dat, aes(xintercept = qrtl), linetype = "solid")+
  geom_vline(data = outcomes_reint_dat, aes(xintercept = fence), linetype = "dashed")+
  geom_rug(alpha = 0.25) +
  facet_wrap(~analysis_plan, ncol = 2) +
  theme_minimal() +
  theme(legend.position = "none", axis.title.y = element_blank(), axis.text.y = element_blank()) +
  labs(x = "Effect size estimate", ); reintegration_outcome_plot

Figure 95: Empirical distribution of effect size estimates, by reintegrational constructs

Mental health outcomes

outcome_mental_health_dat <- 
  mental_health_dat |> 
  group_by(analysis_plan) |> 
  reframe(
    qrtl = quantile(gt,  c(.25, .75)),
    fence = qrtl + 3 * diff(qrtl) * c(-1,1),
    
    .groups = "drop"
  )

mental_health_outcomes_plot <- 
  mental_health_dat |> 
  ggplot(aes(x = gt_pop, fill = analysis_plan)) + 
  geom_density(alpha = 0.7) +
  geom_vline(data = outcome_mental_health_dat, aes(xintercept = qrtl), linetype = "solid")+
  geom_vline(data = outcome_mental_health_dat, aes(xintercept = fence), linetype = "dashed")+
  geom_rug(alpha = 0.25) +
  facet_wrap(~analysis_plan, scales = "free") +
  theme_minimal() +
  theme(legend.position = "none", axis.title.y = element_blank(), axis.text = element_blank()) +
  labs(x = "Effect size estimate"); mental_health_outcomes_plot

Figure 96: Empirical distribution of effect size estimates, by mental health constructs

Colophon

Session Information

─ Session info ───────────────────────────────────────────────────────────────────────────────────
 setting  value
 version  R version 4.5.1 (2025-06-13 ucrt)
 os       Windows 11 x64 (build 22631)
 system   x86_64, mingw32
 ui       RTerm
 language (EN)
 collate  Danish_Denmark.utf8
 ctype    Danish_Denmark.utf8
 tz       Europe/Copenhagen
 date     2025-09-25
 pandoc   3.6.3 @ C:/RStudio-2025.09.0-387/resources/app/bin/quarto/bin/tools/ (via rmarkdown)
 quarto   NA @ C:\\RSTUDI~1.0-3\\RESOUR~1\\app\\bin\\quarto\\bin\\quarto.exe

─ Packages ───────────────────────────────────────────────────────────────────────────────────────
 package      * version    date (UTC) lib source
 base64enc      0.1-3      2015-07-28 [1] CRAN (R 4.5.0)
 cli            3.6.5      2025-04-23 [1] CRAN (R 4.5.1)
 clubSandwich * 0.6.1      2025-07-30 [1] CRAN (R 4.5.1)
 data.table     1.17.8     2025-07-10 [1] CRAN (R 4.5.1)
 digest         0.6.37     2024-08-19 [1] CRAN (R 4.5.1)
 dplyr        * 1.1.4      2023-11-17 [1] CRAN (R 4.5.1)
 evaluate       1.0.5      2025-08-27 [1] CRAN (R 4.5.1)
 farver         2.1.2      2024-05-13 [1] CRAN (R 4.5.1)
 fastDummies  * 1.7.5      2025-01-20 [1] CRAN (R 4.5.1)
 fastmap        1.2.0      2024-05-15 [1] CRAN (R 4.5.1)
 forcats      * 1.0.0      2023-01-29 [1] CRAN (R 4.5.1)
 generics       0.1.4      2025-05-09 [1] CRAN (R 4.5.1)
 GGally       * 2.4.0      2025-08-23 [1] CRAN (R 4.5.1)
 ggExtra      * 0.11.0     2025-09-01 [1] CRAN (R 4.5.1)
 ggh4x        * 0.3.1      2025-05-30 [1] CRAN (R 4.5.1)
 ggplot2      * 4.0.0      2025-09-11 [1] CRAN (R 4.5.1)
 ggrepel      * 0.9.6      2024-09-07 [1] CRAN (R 4.5.1)
 ggridges     * 0.5.7      2025-08-27 [1] CRAN (R 4.5.1)
 ggstats        0.11.0     2025-09-15 [1] CRAN (R 4.5.1)
 glue           1.8.0      2024-09-30 [1] CRAN (R 4.5.1)
 gtable         0.3.6      2024-10-25 [1] CRAN (R 4.5.1)
 hms            1.1.3      2023-03-21 [1] CRAN (R 4.5.1)
 htmltools      0.5.8.1    2024-04-04 [1] CRAN (R 4.5.1)
 htmlwidgets    1.6.4      2023-12-06 [1] CRAN (R 4.5.1)
 httpuv         1.6.16     2025-04-16 [1] CRAN (R 4.5.1)
 igraph       * 2.1.4      2025-01-23 [1] CRAN (R 4.5.1)
 janitor      * 2.2.1      2024-12-22 [1] CRAN (R 4.5.1)
 jsonlite       2.0.0      2025-03-27 [1] CRAN (R 4.5.1)
 kableExtra   * 1.4.0      2024-01-24 [1] CRAN (R 4.5.1)
 knitr        * 1.50       2025-03-16 [1] CRAN (R 4.5.1)
 labeling       0.4.3      2023-08-29 [1] CRAN (R 4.5.0)
 later          1.4.4      2025-08-27 [1] CRAN (R 4.5.1)
 lattice        0.22-7     2025-04-02 [1] CRAN (R 4.5.1)
 lifecycle      1.0.4      2023-11-07 [1] CRAN (R 4.5.1)
 lubridate    * 1.9.4      2024-12-08 [1] CRAN (R 4.5.1)
 magrittr       2.0.4      2025-09-12 [1] CRAN (R 4.5.1)
 mathjaxr       1.8-0      2025-04-30 [1] CRAN (R 4.5.1)
 Matrix       * 1.7-3      2025-03-11 [1] CRAN (R 4.5.1)
 metadat      * 1.4-0      2025-02-04 [1] CRAN (R 4.5.1)
 metafor      * 4.9-18     2025-09-22 [1] Github (wviechtb/metafor@6cc5a0a)
 MetBrewer    * 0.2.0      2022-03-21 [1] CRAN (R 4.5.1)
 mgcv           1.9-3      2025-04-04 [1] CRAN (R 4.5.1)
 mime           0.13       2025-03-17 [1] CRAN (R 4.5.0)
 miniUI         0.1.2      2025-04-17 [1] CRAN (R 4.5.1)
 nlme           3.1-168    2025-03-31 [1] CRAN (R 4.5.1)
 numDeriv     * 2016.8-1.1 2019-06-06 [1] CRAN (R 4.5.0)
 patchwork    * 1.3.2      2025-08-25 [1] CRAN (R 4.5.1)
 pillar         1.11.1     2025-09-17 [1] CRAN (R 4.5.1)
 pkgconfig      2.0.3      2019-09-22 [1] CRAN (R 4.5.1)
 promises       1.3.3      2025-05-29 [1] CRAN (R 4.5.1)
 purrr        * 1.1.0      2025-07-10 [1] CRAN (R 4.5.1)
 R6             2.6.1      2025-02-15 [1] CRAN (R 4.5.1)
 RColorBrewer   1.1-3      2022-04-03 [1] CRAN (R 4.5.0)
 Rcpp           1.1.0      2025-07-02 [1] CRAN (R 4.5.1)
 readr        * 2.1.5      2024-01-10 [1] CRAN (R 4.5.1)
 repr           1.1.7      2024-03-22 [1] CRAN (R 4.5.1)
 rlang        * 1.1.6      2025-04-11 [1] CRAN (R 4.5.1)
 rmarkdown      2.29       2024-11-04 [1] CRAN (R 4.5.1)
 rstudioapi     0.17.1     2024-10-22 [1] CRAN (R 4.5.1)
 S7             0.2.0      2024-11-07 [1] CRAN (R 4.5.1)
 sandwich       3.1-1      2024-09-15 [1] CRAN (R 4.5.1)
 scales         1.4.0      2025-04-24 [1] CRAN (R 4.5.1)
 sessioninfo    1.2.3      2025-02-05 [1] CRAN (R 4.5.1)
 shiny          1.11.1     2025-07-03 [1] CRAN (R 4.5.1)
 skimr        * 2.2.1      2025-07-26 [1] CRAN (R 4.5.1)
 snakecase      0.11.1     2023-08-27 [1] CRAN (R 4.5.1)
 stringi        1.8.7      2025-03-27 [1] CRAN (R 4.5.0)
 stringr      * 1.5.2      2025-09-08 [1] CRAN (R 4.5.1)
 svglite        2.2.1      2025-05-12 [1] CRAN (R 4.5.1)
 systemfonts    1.2.3      2025-04-30 [1] CRAN (R 4.5.1)
 textshaping    1.0.3      2025-09-02 [1] CRAN (R 4.5.1)
 tibble       * 3.3.0      2025-06-08 [1] CRAN (R 4.5.1)
 tidyr        * 1.3.1      2024-01-24 [1] CRAN (R 4.5.1)
 tidyselect     1.2.1      2024-03-11 [1] CRAN (R 4.5.1)
 tidyverse    * 2.0.0      2023-02-22 [1] CRAN (R 4.5.1)
 timechange     0.3.0      2024-01-18 [1] CRAN (R 4.5.1)
 tzdb           0.5.0      2025-03-15 [1] CRAN (R 4.5.1)
 utf8           1.2.6      2025-06-08 [1] CRAN (R 4.5.1)
 vctrs          0.6.5      2023-12-01 [1] CRAN (R 4.5.1)
 viridisLite    0.4.2      2023-05-02 [1] CRAN (R 4.5.1)
 withr          3.0.2      2024-10-28 [1] CRAN (R 4.5.1)
 xfun           0.53       2025-08-19 [1] CRAN (R 4.5.1)
 xml2           1.4.0      2025-08-20 [1] CRAN (R 4.5.1)
 xtable         1.8-4      2019-04-21 [1] CRAN (R 4.5.1)
 yaml           2.3.10     2024-07-26 [1] CRAN (R 4.5.0)
 zoo            1.8-14     2025-04-10 [1] CRAN (R 4.5.1)

 [1] C:/Users/B199526/AppData/Local/Programs/R/R-4.5.1/library
 * ── Packages attached to the search path.

──────────────────────────────────────────────────────────────────────────────────────────────────